geekradar
Int
Explore GeekRadar

More to explore

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

Mistral Large 4

AI model Released

API mode
Chat*
Context window
1,048,576 tokens
Maximum output
262,144 tokens
Open weights
No

Capability & preference

General capability · ECI

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

Mistral Large 4 150.6* Mistral Large 4 · 95% interval 147.4–153.5

Human preference · Arena

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

Mistral Large 4 1430.1 Mistral Large 4 · 95% interval 1423.9–1436.3 · 12,066 votes · 2026-10-28

Model performance

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

Mistral Large 4

Radar task score

45.9* / 100

4/4 core tests · experimental v1

8 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

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

Mistral Large 4 88.1*% Approx. 95% interval: 84.3–91.9%

Novel problem solving

Mystery Game Puzzles · 25%

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

Mistral Large 4 17.0*% Approx. 95% interval: 9.6–24.4%

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

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

Mistral Large 4 58.6*% Approx. 95% interval: 52.9–64.3%

Factual answers

SimpleQA Verified · 25%

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

Mistral Large 4 20.0*% Approx. 95% interval: 17.5–22.5%

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.

Mistral Large 4 98.3*% Approx. 95% interval: 97.1–99.6%

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.

Mistral Large 4 26.8*% Approx. 95% interval: 13.1–40.6%

Chess reasoning

Chess Puzzles v2 · Separate result

Best-move accuracy from text board positions, not images. Only runs after the v2 transition are included; not a chess Elo rating.

Mistral Large 4 33.0*% Approx. 95% interval: 23.7–42.3%

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.

Mistral Large 4 20.8*% Approx. 95% interval: 10.6–31.1%

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.

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

Mistral Large 4

1,491.86 rating

Confidence interval: 1,473.48 – 1,510.25

1055 votes

Category
chinese
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Mistral Large 4

1,478.32 rating

Confidence interval: 1,466.84 – 1,489.81

2770 votes

Category
coding
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Mistral Large 4

1,360.80 rating

Confidence interval: 1,348.59 – 1,373.00

2736 votes

Category
creative_writing
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Mistral Large 4

1,432.47 rating

Confidence interval: 1,423.35 – 1,441.58

4713 votes

Category
english
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Mistral Large 4

1,419.46 rating

Confidence interval: 1,411.09 – 1,427.84

8723 votes

Category
exclude_ties
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Mistral Large 4

1,447.15 rating

Confidence interval: 1,430.43 – 1,463.87

1249 votes

Category
expert
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Mistral Large 4

1,488.57 rating

Confidence interval: 1,460.19 – 1,516.96

483 votes

Category
french
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Mistral Large 4

1,408.95 rating

Confidence interval: 1,366.11 – 1,451.80

180 votes

Category
german
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Mistral Large 4

1,443.68 rating

Confidence interval: 1,436.21 – 1,451.16

7553 votes

Category
hard_prompts
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Mistral Large 4

1,440.90 rating

Confidence interval: 1,429.66 – 1,452.13

3015 votes

Category
hard_prompts_english
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Mistral Large 4

1,438.55 rating

Confidence interval: 1,425.37 – 1,451.74

2060 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Mistral Large 4

1,376.22 rating

Confidence interval: 1,364.82 – 1,387.61

3181 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Mistral Large 4

1,427.58 rating

Confidence interval: 1,408.08 – 1,447.08

974 votes

Category
industry_legal_and_government
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Mistral Large 4

1,442.52 rating

Confidence interval: 1,428.71 – 1,456.32

1958 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Mistral Large 4

1,467.77 rating

Confidence interval: 1,446.28 – 1,489.25

716 votes

Category
industry_mathematical
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Mistral Large 4

1,427.94 rating

Confidence interval: 1,407.66 – 1,448.23

865 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Mistral Large 4

1,467.41 rating

Confidence interval: 1,457.92 – 1,476.90

4177 votes

Category
industry_software_and_it_services
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Mistral Large 4

1,384.67 rating

Confidence interval: 1,373.99 – 1,395.35

3498 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Mistral Large 4

1,418.74 rating

Confidence interval: 1,409.09 – 1,428.39

4230 votes

Category
instruction_following
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Mistral Large 4

1,413.25 rating

Confidence interval: 1,374.49 – 1,452.01

236 votes

Category
japanese
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Mistral Large 4

1,375.14 rating

Confidence interval: 1,331.56 – 1,418.71

186 votes

Category
korean
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Mistral Large 4

1,428.09 rating

Confidence interval: 1,419.06 – 1,437.13

5175 votes

Category
longer_query
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Mistral Large 4

1,467.73 rating

Confidence interval: 1,445.18 – 1,490.27

645 votes

Category
math
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Mistral Large 4

1,437.46 rating

Confidence interval: 1,422.81 – 1,452.11

1773 votes

Category
multi_turn
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Mistral Large 4

1,420.23 rating

Confidence interval: 1,412.69 – 1,427.77

7344 votes

Category
non_english
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Mistral Large 4

1,430.07 rating

Confidence interval: 1,423.88 – 1,436.26

12066 votes

Category
overall
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Mistral Large 4

1,414.43 rating

Confidence interval: 1,399.74 – 1,429.12

1653 votes

Category
russian
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Mistral Large 4

1,470.21 rating

Confidence interval: 1,441.66 – 1,498.77

423 votes

Category
spanish
Evaluated configuration
Mistral Large 4
Evaluation type
human preference
Arena text preference / chinesehuman preference · 2026-10-15 · rating

Mistral Large 4

1,492.00 rating

Confidence interval: 1,469.04 – 1,514.96

658 votes

Category
chinese
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-15 · rating

Mistral Large 4

1,480.89 rating

Confidence interval: 1,466.54 – 1,495.23

1685 votes

Category
coding
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-15 · rating

Mistral Large 4

1,361.95 rating

Confidence interval: 1,346.92 – 1,376.98

1694 votes

Category
creative_writing
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-15 · rating

Mistral Large 4

1,431.86 rating

Confidence interval: 1,420.90 – 1,442.83

2970 votes

Category
english
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-15 · rating

Mistral Large 4

1,418.80 rating

Confidence interval: 1,408.75 – 1,428.84

5352 votes

Category
exclude_ties
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-15 · rating

Mistral Large 4

1,442.23 rating

Confidence interval: 1,421.11 – 1,463.34

754 votes

Category
expert
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-15 · rating

Mistral Large 4

1,494.01 rating

Confidence interval: 1,457.85 – 1,530.18

286 votes

Category
french
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-15 · rating

Mistral Large 4

1,443.56 rating

Confidence interval: 1,434.55 – 1,452.57

4626 votes

Category
hard_prompts
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-15 · rating

Mistral Large 4

1,441.90 rating

Confidence interval: 1,428.04 – 1,455.76

1858 votes

Category
hard_prompts_english
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-15 · rating

Mistral Large 4

1,436.00 rating

Confidence interval: 1,419.52 – 1,452.48

1233 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-15 · rating

Mistral Large 4

1,376.21 rating

Confidence interval: 1,362.19 – 1,390.22

1960 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-15 · rating

Mistral Large 4

1,422.27 rating

Confidence interval: 1,398.36 – 1,446.17

597 votes

Category
industry_legal_and_government
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-15 · rating

Mistral Large 4

1,444.29 rating

Confidence interval: 1,427.38 – 1,461.20

1232 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-15 · rating

Mistral Large 4

1,460.52 rating

Confidence interval: 1,431.84 – 1,489.20

393 votes

Category
industry_mathematical
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-15 · rating

Mistral Large 4

1,422.69 rating

Confidence interval: 1,396.86 – 1,448.52

523 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-15 · rating

Mistral Large 4

1,468.13 rating

Confidence interval: 1,456.64 – 1,479.63

2593 votes

Category
industry_software_and_it_services
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-15 · rating

Mistral Large 4

1,383.18 rating

Confidence interval: 1,370.04 – 1,396.31

2139 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-15 · rating

Mistral Large 4

1,420.64 rating

Confidence interval: 1,408.83 – 1,432.45

2579 votes

Category
instruction_following
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-15 · rating

Mistral Large 4

1,426.76 rating

Confidence interval: 1,415.75 – 1,437.76

3141 votes

Category
longer_query
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-15 · rating

Mistral Large 4

1,469.19 rating

Confidence interval: 1,440.10 – 1,498.28

377 votes

Category
math
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-15 · rating

Mistral Large 4

1,437.29 rating

Confidence interval: 1,419.04 – 1,455.54

1069 votes

Category
multi_turn
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-15 · rating

Mistral Large 4

1,420.16 rating

Confidence interval: 1,411.04 – 1,429.29

4479 votes

Category
non_english
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-15 · rating

Mistral Large 4

1,429.99 rating

Confidence interval: 1,422.78 – 1,437.21

7451 votes

Category
overall
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-15 · rating

Mistral Large 4

1,421.99 rating

Confidence interval: 1,403.28 – 1,440.69

953 votes

Category
russian
Evaluated configuration
mistral-large-4
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-15 · rating

Mistral Large 4

1,443.07 rating

Confidence interval: 1,405.39 – 1,480.76

249 votes

Category
spanish
Evaluated configuration
mistral-large-4
Evaluation type
human preference

Run it locally

Inference · memory estimate

Mistral Large 4

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.

Mistral Large 4

Mistral Large 4 / Pricing tiers 1
Provider / scopeRateCurrencyAmount / unit
MistralMistral Large 4Standard Cached inputUSD0.07 / 1M tokens
InputUSD0.7 / 1M tokens
OutputUSD2.1 / 1M tokens

Capabilities

Context window
1,048,576 tokens
Maximum input
1,048,576 tokens*
Maximum output
262,144 tokens
Reasoning
Yes
Tool calling
Yes
Structured output
Yes*
Open weights
No
API mode
Chat*
Image input
Yes*
Prompt caching
Yes*
Tool choice
Yes*
Model license
Proprietary

Input modalities

  • text
  • image

Output modalities

  • text

System configuration

Detailed specifications

Specifications · 15 fields
cache read input token cost
6.8e-8*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
6.8e-7*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
1,048,576*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
1,048,576*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
2.09e-6*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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.*
reasoning effort levels
  • none*
  • high*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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 assistant prefill
Yes*
Scope and conditions
native_context
provider
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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
mistral*
mode
chat*
full_native_model_id
mistral/mistral-large-4*
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.*

* Provisional data