geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
openai

GPT-5.6 Sol

AI model Released

API mode
Chat*
Context window
1,050,000 tokens
Maximum output
128,000 tokens
Open weights
No

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

GPT-5.6 Sol 162.6* GPT-5.6 Sol · 95% interval 159.6–166.2

Human preference · Arena

Overall text preferences from head-to-head votes. Not factual accuracy or a task-success rate.

GPT-5.6 Sol 1457.2 GPT-5.6 Sol (xHigh) · 95% interval 1453.0–1461.5 · 43,796 votes · 2026-10-28

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

GPT-5.6 Sol

Radar task score

77.6* / 100

4/4 core tests · experimental v1

7 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

GPT-5.6 Sol 93.5*% Approx. 95% interval: 90.4–96.6%

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

GPT-5.6 Sol 58.0*% Approx. 95% interval: 48.3–67.7%

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

GPT-5.6 Sol 89.1*% Approx. 95% interval: 85.5–92.7%

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

GPT-5.6 Sol 69.7*% Approx. 95% interval: 66.9–72.5%

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

GPT-5.6 Sol 100.0*% Uncertainty not available

Research mathematics

FrontierMath Tier 4 v2 · Separate result

The hardest private FrontierMath problems, corrected v2 set, with Python tools. Separate from Tiers 1-3 in the core score.

GPT-5.6 Sol 82.9*% Approx. 95% interval: 71.3–94.6%

Visual spatial reasoning

Furniture Assembly · Separate result

Assembly-error detection from photos and manuals, with zoom and Python tools. Uses model-assisted grading; not a general vision score.

GPT-5.6 Sol 56.7*% Approx. 95% interval: 44.2–69.1%

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Benchmark results

  • $OneMillion-Bench (expert score) · version unreported · reported score · max53.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Judge
    gemini-3.1-pro-preview*
  • Agents' Last Exam (Pass / Score) · version unreported · Pass · max30.60*Platform unreported Compare results
    Test conditions
    Metric
    Pass*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • Agents' Last Exam (Pass / Score) · version unreported · Score · max53.60*Platform unreported Compare results
    Test conditions
    Metric
    Score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • AndroidBench · version unreported · reported score · Public subset · max · avg@374*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Tasks
    95*
    Statistic
    avg@3*
    Task set
    Public subset*
  • Automation-Bench (Pass@1) · version unreported · reported score · Public subset · max29.70*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Tasks
    600*
    Task set
    Public subset*
  • CoWorkBench · version unreported · reported score · Internal · max71.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Task set
    Internal*
  • DeepSWE 1.1 · 1.1 · reported score · Claude Code or mini-SWE-agent; winning harness unspecified · max · Highest result across two harnesses73*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Claude Code or mini-SWE-agent; winning harness unspecified*
    Statistic
    Highest result across two harnesses*
    Temperature
    1.0*
    Top p
    0.95*
    Context length
    256K tokens*
  • GPQA Diamond · version unreported · reported score · max94.10*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • HealthBench · version unreported · reported score · max55.30*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • HLE · version unreported · reported score · Disabled · max47.20*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Tool access
    Disabled*
  • HLE w/ tools · version unreported · reported score · Enabled · max58*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Tool access
    Enabled*
  • IFBench · version unreported · reported score · max72.70*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • JobBench · version unreported · reported score · max45.40*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • LongBench v2 · 2 · reported score · max67.10*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • MLS-Bench-Lite · version unreported · reported score · Not specified · max46.20*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Not specified*
  • MRCR v2 256K (8-needle) · 2 · reported score · max93.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Variant
    8-needle*
    Context length
    256K tokens*
  • PaperBench · version unreported · reported score · BasicAgent · max90.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    BasicAgent*
    Variant
    Code-Dev*
    Judge
    Claude Opus 4.6*
    Runs
    3*
    Timeout
    12 hours per run*
  • PLawBench · version unreported · reported score · max72.30*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Judge
    gemini-3.1-pro-preview*
  • PRBench-Finance · version unreported · reported score · max55.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • PRBench-Legal · version unreported · reported score · max57.60*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • QwenQoderBench · version unreported · reported score · Claude Code · Internal · max · avg@553.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Claude Code*
    Statistic
    avg@5*
    Timeout
    6 hours*
    Task set
    Internal*
    Maximum output
    32,768*
    Temperature
    1.0*
    Context length
    256K tokens*
  • QwenReactBench · version unreported · reported score · Claude Code · Internal · max1,564* EloPlatform unreported Compare results
    Test conditions
    Metric
    BT/Elo rating*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Languages
    English and Chinese*
    Judge
    Multimodal automatic judge*
    Task set
    Internal*
    Harness
    Claude Code*
  • QwenSVGBench · version unreported · reported score · Not specified · Internal · max1,758* EloPlatform unreported Compare results
    Test conditions
    Metric
    BT/Elo rating*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Languages
    English and Chinese*
    Judge
    Multimodal automatic judge*
    Task set
    Internal*
    Harness
    Not specified*
  • QwenSWEBench · version unreported · reported score · Claude Code · Internal · max · avg@373.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Claude Code*
    Statistic
    avg@3*
    Timeout
    8 hours*
    Task set
    Internal*
    Maximum output
    32,768*
    Temperature
    1.0*
    Context length
    256K tokens*
  • SkillsBench · version unreported · reported score · Codex · Public SkillsBench v1.1 · max · Average of three runs per task73.50*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Codex*
    Tasks
    87*
    Statistic
    Average of three runs per task*
    Task set
    Public SkillsBench v1.1*
  • Reported score · version unreported · intelligence59*Platform unreported Compare results
  • SWE-bench Pro · version unreported · reported score · Claude Code · Refined task set with corrected problematic tasks · max64.60*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Claude Code*
    Task set
    Refined task set with corrected problematic tasks*
    Temperature
    1.0*
    Top p
    0.95*
    Context length
    256K tokens*
  • Terminal Bench 2.1 · 2.1 · reported score · Codex · max · Best published result across harnesses88.80*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
    Harness
    Codex*
    Statistic
    Best published result across harnesses*
  • Toolathlon Verified (Pass@1) · version unreported · reported score · max74.90*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*
  • WorkSpaceBench · version unreported · reported score · max65.60*Platform unreported Compare results
    Test conditions
    Metric
    reported score*
    Checkpoint version
    Not specified*
    Reasoning setting
    max*

Recorded results

Only the same benchmark, version, workload, scenario and unit share a row. Missing results are not zero. System results include the full submitted hardware and software configuration.

Arena text preference / chinesehuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,525.59 rating

Confidence interval: 1,514.70 – 1,536.48

3417 votes

Category
chinese
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,495.18 rating

Confidence interval: 1,488.66 – 1,501.70

11636 votes

Category
coding
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,450.28 rating

Confidence interval: 1,442.86 – 1,457.69

9174 votes

Category
creative_writing
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,458.15 rating

Confidence interval: 1,452.53 – 1,463.76

17540 votes

Category
english
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,458.57 rating

Confidence interval: 1,453.15 – 1,463.99

32891 votes

Category
exclude_ties
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,515.73 rating

Confidence interval: 1,506.62 – 1,524.84

4879 votes

Category
expert
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,473.06 rating

Confidence interval: 1,456.01 – 1,490.10

1461 votes

Category
french
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,473.46 rating

Confidence interval: 1,450.74 – 1,496.17

693 votes

Category
german
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,483.14 rating

Confidence interval: 1,478.21 – 1,488.08

28350 votes

Category
hard_prompts
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,477.11 rating

Confidence interval: 1,470.59 – 1,483.63

11370 votes

Category
hard_prompts_english
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,456.21 rating

Confidence interval: 1,448.80 – 1,463.61

8026 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,449.56 rating

Confidence interval: 1,442.71 – 1,456.40

11155 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,473.12 rating

Confidence interval: 1,462.70 – 1,483.54

3656 votes

Category
industry_legal_and_government
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,455.28 rating

Confidence interval: 1,447.54 – 1,463.02

7111 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,493.76 rating

Confidence interval: 1,481.21 – 1,506.31

2461 votes

Category
industry_mathematical
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,429.78 rating

Confidence interval: 1,418.44 – 1,441.12

3145 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,487.95 rating

Confidence interval: 1,482.24 – 1,493.66

16872 votes

Category
industry_software_and_it_services
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,463.55 rating

Confidence interval: 1,456.91 – 1,470.20

11965 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,482.76 rating

Confidence interval: 1,476.78 – 1,488.75

15280 votes

Category
instruction_following
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,472.08 rating

Confidence interval: 1,447.92 – 1,496.24

711 votes

Category
japanese
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,441.70 rating

Confidence interval: 1,420.15 – 1,463.25

838 votes

Category
korean
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,480.00 rating

Confidence interval: 1,474.37 – 1,485.64

20081 votes

Category
longer_query
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,472.14 rating

Confidence interval: 1,459.09 – 1,485.19

2112 votes

Category
math
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,461.22 rating

Confidence interval: 1,453.38 – 1,469.06

7135 votes

Category
multi_turn
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,450.36 rating

Confidence interval: 1,445.36 – 1,455.35

26248 votes

Category
non_english
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,457.24 rating

Confidence interval: 1,453.01 – 1,461.47

43796 votes

Category
overall
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,452.99 rating

Confidence interval: 1,430.12 – 1,475.86

658 votes

Category
polish
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,471.63 rating

Confidence interval: 1,462.78 – 1,480.49

5206 votes

Category
russian
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

GPT-5.6 Sol

1,442.69 rating

Confidence interval: 1,425.81 – 1,459.57

1302 votes

Category
spanish
Evaluated configuration
GPT-5.6 Sol (xHigh)
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,526.63 rating

Confidence interval: 1,515.31 – 1,537.96

3113 votes

Category
chinese
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,496.47 rating

Confidence interval: 1,489.81 – 1,503.14

10825 votes

Category
coding
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,449.01 rating

Confidence interval: 1,441.40 – 1,456.62

8408 votes

Category
creative_writing
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,457.95 rating

Confidence interval: 1,452.23 – 1,463.67

16314 votes

Category
english
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,458.48 rating

Confidence interval: 1,452.94 – 1,464.03

30398 votes

Category
exclude_ties
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,516.23 rating

Confidence interval: 1,506.83 – 1,525.64

4517 votes

Category
expert
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,474.46 rating

Confidence interval: 1,456.59 – 1,492.33

1298 votes

Category
french
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,475.50 rating

Confidence interval: 1,452.13 – 1,498.87

644 votes

Category
german
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,483.33 rating

Confidence interval: 1,478.30 – 1,488.35

26312 votes

Category
hard_prompts
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,476.85 rating

Confidence interval: 1,470.20 – 1,483.50

10621 votes

Category
hard_prompts_english
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,455.98 rating

Confidence interval: 1,448.42 – 1,463.53

7516 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,450.15 rating

Confidence interval: 1,443.15 – 1,457.15

10300 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,476.10 rating

Confidence interval: 1,465.40 – 1,486.80

3432 votes

Category
industry_legal_and_government
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,455.67 rating

Confidence interval: 1,447.73 – 1,463.61

6582 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,495.34 rating

Confidence interval: 1,482.26 – 1,508.41

2244 votes

Category
industry_mathematical
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,431.91 rating

Confidence interval: 1,420.20 – 1,443.61

2922 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,489.22 rating

Confidence interval: 1,483.41 – 1,495.04

15703 votes

Category
industry_software_and_it_services
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,462.61 rating

Confidence interval: 1,455.80 – 1,469.41

10968 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,482.31 rating

Confidence interval: 1,476.18 – 1,488.44

14119 votes

Category
instruction_following
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,470.91 rating

Confidence interval: 1,444.85 – 1,496.97

623 votes

Category
japanese
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,442.12 rating

Confidence interval: 1,419.97 – 1,464.26

787 votes

Category
korean
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,480.81 rating

Confidence interval: 1,475.07 – 1,486.56

18670 votes

Category
longer_query
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,473.20 rating

Confidence interval: 1,459.54 – 1,486.87

1902 votes

Category
math
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,459.89 rating

Confidence interval: 1,451.84 – 1,467.93

6597 votes

Category
multi_turn
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,450.83 rating

Confidence interval: 1,445.72 – 1,455.93

24121 votes

Category
non_english
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,457.06 rating

Confidence interval: 1,452.73 – 1,461.39

40442 votes

Category
overall
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,461.39 rating

Confidence interval: 1,437.73 – 1,485.06

616 votes

Category
polish
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,470.37 rating

Confidence interval: 1,461.22 – 1,479.52

4774 votes

Category
russian
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

GPT-5.6 Sol

1,440.59 rating

Confidence interval: 1,422.92 – 1,458.26

1179 votes

Category
spanish
Evaluated configuration
gpt-5.6-sol-xhigh
Evaluation type
human preference

Run it locally

Inference · memory estimate

GPT-5.6 Sol

Closed weights. No supported local deployment or GPU requirement is published.

API cost

USD / month

Standard text-token estimate. Excludes taxes, caching, batch discounts, reasoning and multimodal surcharges. Provider limits and regional availability may differ.

GPT-5.6 Sol

GPT-5.6 Sol / Pricing tiers 8
Provider / scopeRateCurrencyAmount / unit
OpenAIgpt-5.6-solBatch Cached inputUSD0.2 / 1M tokens
InputUSD2 / 1M tokens
OutputUSD10 / 1M tokens
OpenAIgpt-5.6-solBatch long_context Cached inputUSD0.4 / 1M tokens
InputUSD4 / 1M tokens
OutputUSD15 / 1M tokens
OpenAIgpt-5.6-solFlex Cached inputUSD0.2 / 1M tokens
InputUSD2 / 1M tokens
OutputUSD10 / 1M tokens
OpenAIgpt-5.6-solFlex long_context Cached inputUSD0.4 / 1M tokens
InputUSD4 / 1M tokens
OutputUSD15 / 1M tokens
OpenAIgpt-5.6-solPriority long_context Cached inputUSD1.6 / 1M tokens
InputUSD16 / 1M tokens
OutputUSD60 / 1M tokens
OpenAIgpt-5.6-solPriority Cached inputUSD0.8 / 1M tokens
InputUSD8 / 1M tokens
OutputUSD40 / 1M tokens
OpenAIgpt-5.6-solStandard long_context Cached inputUSD0.8 / 1M tokens
InputUSD8 / 1M tokens
OutputUSD30 / 1M tokens
OpenAIgpt-5.6-solStandard Cached inputUSD0.4 / 1M tokens
InputUSD4 / 1M tokens
OutputUSD20 / 1M tokens

Capabilities

Context window
1,050,000 tokens
Maximum input
922,000 tokens*
Maximum output
128,000 tokens
Reasoning
Yes
Tool calling
Yes
Structured output
Yes*
Open weights
No
Knowledge cutoff
2026-02-16
API mode
Chat*
Declared context window
1.05M*
Model type
Reasoning / agentic*
Declared modalities
Text, vision*
Image input
Yes*
PDF input
Yes*
Parallel tool calls
Yes*
Prompt caching
Yes*
System messages
Yes*
Tool choice
Yes*
Web search
Yes*
Model license
Proprietary

Input modalities

  • text
  • image
  • pdf

Output modalities

  • text

System configuration

Detailed specifications

Specifications 1 · 13 fields
api providers
OpenAI API, Microsoft Azure AI*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
context window
1.05M*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
developer
OpenAI*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
input price
5*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
intelligence
59*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
license
Proprietary*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
max output
128K*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
modality
Text, vision*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
model type
Reasoning / agentic*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
name
GPT-5.6 Sol*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
open weights
no*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
output price
30*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
release date
2026-06*
Scope and conditions
native_context
metadata
name
Convly AI Models Database*
description
Hand-verified specifications, licences, API pricing (USD) and local-hardware VRAM requirements for frontier and open-weight AI models.*
generated
2026-09-22T15:41:48+00:00*
model_count
43*
fields
  • slug*
  • name*
  • developer*
  • model_type*
  • modality*
  • parameters*
  • context_window*
  • max_output*
  • open_weights*
  • license*
  • release_date*
  • input_price*
  • output_price*
  • price_unit*
  • price_tiers*
  • vram_q4*
  • vram_q4_gb*
  • min_gpu*
  • intelligence*
  • api_providers*
  • official_url*
  • retired*
  • convly_url*
developer
OpenAI*
price_unit
Not reported*
price_tiers
Not reported*
Specifications 2 · 60 fields
cache creation input token cost
5.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost above 272k tokens
1.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost above 272k tokens batches
5.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost above 272k tokens flex
5.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost above 272k tokens priority
2.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost batches
2.5e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost flex
2.5e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache creation input token cost priority
1.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost
4.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost above 272k tokens
8.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost above 272k tokens batches
4.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost above 272k tokens flex
4.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost above 272k tokens priority
1.6e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost batches
2.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost flex
2.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
cache read input token cost priority
8.0e-7*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token
4.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token above 272k tokens
8.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token above 272k tokens batches
4.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token above 272k tokens flex
4.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token above 272k tokens priority
1.6e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token batches
2.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token flex
2.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
input cost per token priority
8.0e-6*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
litellm provider
openai*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
max input tokens
922,000*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
max output tokens
128,000*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
max tokens
128,000*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
mode
chat*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token
2.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token above 272k tokens
3.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token above 272k tokens batches
1.5e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token above 272k tokens flex
1.5e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token above 272k tokens priority
6.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token batches
1.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token flex
1.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
output cost per token priority
4.0e-5*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
regional processing uplift multiplier eu
1.1*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
regional processing uplift multiplier us
1.1*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
search context cost per query
search_context_size_high
0.01*
search_context_size_low
0.01*
search_context_size_medium
0.01*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supported endpoints
  • /v1/chat/completions*
  • /v1/batch*
  • /v1/responses*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supported modalities
  • text*
  • image*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supported output modalities
  • text*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports computer use
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports function calling
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports minimal reasoning effort
No*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports native streaming
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports none reasoning effort
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports parallel function calling
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports pdf input
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports prompt cache breakpoint
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports prompt caching
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports reasoning
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports response schema
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports system messages
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports tool choice
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports tool search
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports vision
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports web search
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
supports xhigh reasoning effort
Yes*
Scope and conditions
native_context
provider
openai*
mode
chat*
full_native_model_id
gpt-5.6-sol*
pricing_scope
Literal source field names and values retained; provider, region, cache/batch/threshold conditions and per-unit meaning must be checked. max_tokens is not automatically a context limit.*
Specifications 3 · 29 fields
QwenSWEBench
suite
QwenSWEBench
workload
reported score · Claude Code · Internal · max · avg@3
score
73.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Claude Code*
Statistic
avg@3*
Timeout
8 hours*
Task set
Internal*
Maximum output
32,768*
Temperature
1.0*
Context length
256K tokens*
QwenQoderBench
suite
QwenQoderBench
workload
reported score · Claude Code · Internal · max · avg@5
score
53.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Claude Code*
Statistic
avg@5*
Timeout
6 hours*
Task set
Internal*
Maximum output
32,768*
Temperature
1.0*
Context length
256K tokens*
QwenReactBench
suite
QwenReactBench
workload
reported score · Claude Code · Internal · max
score
1,564.0*
unit
Elo
Scope and conditions
Test conditions
Metric
BT/Elo rating*
Checkpoint version
Not specified*
Reasoning setting
max*
Languages
English and Chinese*
Judge
Multimodal automatic judge*
Task set
Internal*
Harness
Claude Code*
QwenSVGBench
suite
QwenSVGBench
workload
reported score · Not specified · Internal · max
score
1,758.0*
unit
Elo
Scope and conditions
Test conditions
Metric
BT/Elo rating*
Checkpoint version
Not specified*
Reasoning setting
max*
Languages
English and Chinese*
Judge
Multimodal automatic judge*
Task set
Internal*
Harness
Not specified*
CoWorkBench
suite
CoWorkBench
workload
reported score · Internal · max
score
71.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Task set
Internal*
WorkSpaceBench
suite
WorkSpaceBench
workload
reported score · max
score
65.6*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
JobBench
suite
JobBench
workload
reported score · max
score
45.4*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
SkillsBench
suite
SkillsBench
workload
reported score · Codex · Public SkillsBench v1.1 · max · Average of three runs per task
score
73.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Codex*
Tasks
87*
Statistic
Average of three runs per task*
Task set
Public SkillsBench v1.1*
Agents' Last Exam (Pass / Score)
suite
Agents' Last Exam (Pass / Score)
workload
Pass · max
score
30.6*
Scope and conditions
Test conditions
Metric
Pass*
Checkpoint version
Not specified*
Reasoning setting
max*
Agents' Last Exam (Pass / Score)
suite
Agents' Last Exam (Pass / Score)
workload
Score · max
score
53.6*
Scope and conditions
Test conditions
Metric
Score*
Checkpoint version
Not specified*
Reasoning setting
max*
Terminal Bench 2.1
suite
Terminal Bench 2.1
version
2.1
workload
reported score · Codex · max · Best published result across harnesses
score
88.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Codex*
Statistic
Best published result across harnesses*
Automation-Bench (Pass@1)
suite
Automation-Bench (Pass@1)
workload
reported score · Public subset · max
score
29.7*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Tasks
600*
Task set
Public subset*
Toolathlon Verified (Pass@1)
suite
Toolathlon Verified (Pass@1)
workload
reported score · max
score
74.9*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
HLE w/ tools
suite
HLE w/ tools
workload
reported score · Enabled · max
score
58.0*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Tool access
Enabled*
GPQA Diamond
suite
GPQA Diamond
workload
reported score · max
score
94.1*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
HLE
suite
HLE
workload
reported score · Disabled · max
score
47.2*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Tool access
Disabled*
IFBench
suite
IFBench
workload
reported score · max
score
72.7*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
$OneMillion-Bench (expert score)
suite
$OneMillion-Bench (expert score)
workload
reported score · max
score
53.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Judge
gemini-3.1-pro-preview*
HealthBench
suite
HealthBench
workload
reported score · max
score
55.3*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
SWE-bench Pro
suite
SWE-bench Pro
workload
reported score · Claude Code · Refined task set with corrected problematic tasks · max
score
64.6*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Claude Code*
Task set
Refined task set with corrected problematic tasks*
Temperature
1.0*
Top p
0.95*
Context length
256K tokens*
PLawBench
suite
PLawBench
workload
reported score · max
score
72.3*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Judge
gemini-3.1-pro-preview*
PRBench-Legal
suite
PRBench-Legal
workload
reported score · max
score
57.6*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
PRBench-Finance
suite
PRBench-Finance
workload
reported score · max
score
55.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
MRCR v2 256K (8-needle)
suite
MRCR v2 256K (8-needle)
version
2
workload
reported score · max
score
93.8*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Variant
8-needle*
Context length
256K tokens*
LongBench v2
suite
LongBench v2
version
2
workload
reported score · max
score
67.1*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
DeepSWE 1.1
suite
DeepSWE 1.1
version
1.1
workload
reported score · Claude Code or mini-SWE-agent; winning harness unspecified · max · Highest result across two harnesses
score
73.0*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Claude Code or mini-SWE-agent; winning harness unspecified*
Statistic
Highest result across two harnesses*
Temperature
1.0*
Top p
0.95*
Context length
256K tokens*
MLS-Bench-Lite
suite
MLS-Bench-Lite
workload
reported score · Not specified · max
score
46.2*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
Not specified*
PaperBench
suite
PaperBench
workload
reported score · BasicAgent · max
score
90.5*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Harness
BasicAgent*
Variant
Code-Dev*
Judge
Claude Opus 4.6*
Runs
3*
Timeout
12 hours per run*
AndroidBench
suite
AndroidBench
workload
reported score · Public subset · max · avg@3
score
74.0*
Scope and conditions
Test conditions
Metric
reported score*
Checkpoint version
Not specified*
Reasoning setting
max*
Tasks
95*
Statistic
avg@3*
Task set
Public subset*

* Provisional data