geekradar
Int
Explore GeekRadar

More to explore

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

Kimi K3

AI model Released

API mode
Chat*
Context window
1,048,576 tokens
Maximum output
131,072 tokens
Open weights
Yes

Capability & preference

General capability · ECI

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

Kimi K3 158.3* Kimi K3 · 95% interval 155.8–161.4

Human preference · Arena

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

Kimi K3 1475.7 Kimi K3 (Max) · 95% interval 1471.3–1480.1 · 36,780 votes · 2026-10-28

Model performance

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

Kimi K3

Radar task score

60.5* / 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%.

Kimi K3 93.1*% Approx. 95% interval: 90.2–96.0%

Novel problem solving

Mystery Game Puzzles · 25%

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

Kimi K3 26.0*% Approx. 95% interval: 17.4–34.6%

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

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

Kimi K3 72.2*% Approx. 95% interval: 67.0–77.4%

Factual answers

SimpleQA Verified · 25%

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

Kimi K3 50.6*% Approx. 95% interval: 47.5–53.7%

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.

Kimi K3 97.2*% Approx. 95% interval: 95.0–99.4%

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.

Kimi K3 39.0*% Approx. 95% interval: 23.9–54.1%

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.

Kimi K3 34.2*% Approx. 95% interval: 22.2–46.2%

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

  • Reported score · version unreported · intelligence57*Platform unreported Compare results

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

Kimi K3

1,530.58 rating

Confidence interval: 1,518.70 – 1,542.47

2688 votes

Category
chinese
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / codinghuman preference · 2026-10-28 · rating

Kimi K3

1,508.38 rating

Confidence interval: 1,501.43 – 1,515.34

9235 votes

Category
coding
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / creative_writinghuman preference · 2026-10-28 · rating

Kimi K3

1,457.17 rating

Confidence interval: 1,449.28 – 1,465.05

7452 votes

Category
creative_writing
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / englishhuman preference · 2026-10-28 · rating

Kimi K3

1,481.17 rating

Confidence interval: 1,475.27 – 1,487.07

13907 votes

Category
english
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / exclude_tieshuman preference · 2026-10-28 · rating

Kimi K3

1,483.46 rating

Confidence interval: 1,477.74 – 1,489.18

27079 votes

Category
exclude_ties
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / experthuman preference · 2026-10-28 · rating

Kimi K3

1,519.85 rating

Confidence interval: 1,509.28 – 1,530.43

3347 votes

Category
expert
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / frenchhuman preference · 2026-10-28 · rating

Kimi K3

1,486.06 rating

Confidence interval: 1,468.87 – 1,503.25

1434 votes

Category
french
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / germanhuman preference · 2026-10-28 · rating

Kimi K3

1,489.47 rating

Confidence interval: 1,466.63 – 1,512.31

712 votes

Category
german
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / hard_promptshuman preference · 2026-10-28 · rating

Kimi K3

1,495.54 rating

Confidence interval: 1,490.43 – 1,500.65

23477 votes

Category
hard_prompts
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / hard_prompts_englishhuman preference · 2026-10-28 · rating

Kimi K3

1,496.12 rating

Confidence interval: 1,489.11 – 1,503.12

8930 votes

Category
hard_prompts_english
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_business_and_management_and_financial_operationshuman preference · 2026-10-28 · rating

Kimi K3

1,468.88 rating

Confidence interval: 1,461.13 – 1,476.64

6887 votes

Category
industry_business_and_management_and_financial_operations
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_entertainment_and_sports_and_mediahuman preference · 2026-10-28 · rating

Kimi K3

1,452.21 rating

Confidence interval: 1,444.95 – 1,459.47

9062 votes

Category
industry_entertainment_and_sports_and_media
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_legal_and_governmenthuman preference · 2026-10-28 · rating

Kimi K3

1,501.70 rating

Confidence interval: 1,490.09 – 1,513.30

2859 votes

Category
industry_legal_and_government
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_life_and_physical_and_social_sciencehuman preference · 2026-10-28 · rating

Kimi K3

1,507.74 rating

Confidence interval: 1,499.42 – 1,516.06

5732 votes

Category
industry_life_and_physical_and_social_science
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_mathematicalhuman preference · 2026-10-28 · rating

Kimi K3

1,504.02 rating

Confidence interval: 1,490.18 – 1,517.86

1943 votes

Category
industry_mathematical
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_medicine_and_healthcarehuman preference · 2026-10-28 · rating

Kimi K3

1,487.29 rating

Confidence interval: 1,474.92 – 1,499.66

2489 votes

Category
industry_medicine_and_healthcare
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_software_and_it_serviceshuman preference · 2026-10-28 · rating

Kimi K3

1,498.59 rating

Confidence interval: 1,492.63 – 1,504.54

13902 votes

Category
industry_software_and_it_services
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / industry_writing_and_literature_and_languagehuman preference · 2026-10-28 · rating

Kimi K3

1,471.11 rating

Confidence interval: 1,464.08 – 1,478.14

9815 votes

Category
industry_writing_and_literature_and_language
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / instruction_followinghuman preference · 2026-10-28 · rating

Kimi K3

1,485.25 rating

Confidence interval: 1,478.94 – 1,491.55

12229 votes

Category
instruction_following
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / japanesehuman preference · 2026-10-28 · rating

Kimi K3

1,495.24 rating

Confidence interval: 1,471.16 – 1,519.32

725 votes

Category
japanese
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / koreanhuman preference · 2026-10-28 · rating

Kimi K3

1,458.56 rating

Confidence interval: 1,436.64 – 1,480.49

769 votes

Category
korean
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / longer_queryhuman preference · 2026-10-28 · rating

Kimi K3

1,493.48 rating

Confidence interval: 1,487.49 – 1,499.48

15841 votes

Category
longer_query
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / mathhuman preference · 2026-10-28 · rating

Kimi K3

1,492.43 rating

Confidence interval: 1,477.91 – 1,506.95

1673 votes

Category
math
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / multi_turnhuman preference · 2026-10-28 · rating

Kimi K3

1,489.57 rating

Confidence interval: 1,481.04 – 1,498.11

5628 votes

Category
multi_turn
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / non_englishhuman preference · 2026-10-28 · rating

Kimi K3

1,465.93 rating

Confidence interval: 1,460.77 – 1,471.08

22861 votes

Category
non_english
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / overallhuman preference · 2026-10-28 · rating

Kimi K3

1,475.69 rating

Confidence interval: 1,471.25 – 1,480.12

36780 votes

Category
overall
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / polishhuman preference · 2026-10-28 · rating

Kimi K3

1,462.12 rating

Confidence interval: 1,438.90 – 1,485.34

620 votes

Category
polish
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / russianhuman preference · 2026-10-28 · rating

Kimi K3

1,484.03 rating

Confidence interval: 1,474.84 – 1,493.23

4663 votes

Category
russian
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference
Arena text preference / spanishhuman preference · 2026-10-28 · rating

Kimi K3

1,477.22 rating

Confidence interval: 1,459.45 – 1,494.99

1158 votes

Category
spanish
Evaluated configuration
Kimi K3 (Max)
Evaluation type
human preference

Run it locally

Inference · memory estimate

Kimi K3

Not enough verified architecture information for a reliable VRAM estimate.

API cost

USD / month

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

Kimi K3

Kimi K3 / Pricing tiers 4
Provider / scopeRateCurrencyAmount / unit
DeepInframoonshotai/Kimi-K3Standard Cached inputUSD0.285 / 1M tokens
InputUSD2.85 / 1M tokens
OutputUSD14.25 / 1M tokens
Fireworks AIKimi K3Priority Cached inputUSD0.375 / 1M tokens
InputUSD3.75 / 1M tokens
OutputUSD18.75 / 1M tokens
Fireworks AIKimi K3Standard Cached inputUSD0.3 / 1M tokens
InputUSD3 / 1M tokens
OutputUSD15 / 1M tokens
Fireworks AIKimi K3 FastStandard Cached inputUSD0.45 / 1M tokens
InputUSD4.5 / 1M tokens
OutputUSD22.5 / 1M tokens

Capabilities

Context window
1,048,576 tokens
Maximum input
1,000,000 tokens*
Maximum output
131,072 tokens
Reasoning
Yes
Tool calling
Yes
Structured output
Yes*
Open weights
Yes
API mode
Chat*
Declared context window
1M*
Model type
LLM (MoE, reasoning, multimodal)*
Declared modalities
Text, Vision → Text*
Image input
Yes*
Audio input
No*
Prompt caching
Yes*
System messages
Yes*
Tool choice
Yes*
Model license
Open weights (due 2026-07-27)*

Input modalities

  • text
  • image
  • video

Output modalities

  • text

Processing & platform

Architecture
kimi_k3

System configuration

Detailed specifications

Specifications 1 · 15 fields
api providers
Moonshot (Kimi), OpenRouter*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
context window
1M*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
developer
Moonshot 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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
input price
3*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
intelligence
57*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
license
Open weights (due 2026-07-27)*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
min gpu
Multi-node cluster (e.g. 2× 8-GPU H200 141GB nodes)*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
modality
Text, Vision → Text*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
model type
LLM (MoE, reasoning, multimodal)*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
name
Kimi K3*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
open weights
yes*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
output price
15*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
parameters
2.8T total / 16 of 896 experts active (MoE)*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
release date
2026-07*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
vram q4
~1.4 TB*
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
Moonshot AI*
price_unit
Not reported*
price_tiers
Not reported*
Specifications 2 · 29 fields
cache creation input token cost
4.125e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
2.0625e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
7.21875e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
3.3e-7*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
1.65e-7*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
5.775e-7*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
3.3e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
1.65e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
5.775e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
1,000,000*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
131,072*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
131,072*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
1.65e-5*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
8.25e-6*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
2.8875e-5*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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 audio input
No*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
No*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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 regex lookaround
No*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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.*
token count framing
64*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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.*
token count tokenizer
kimi_k3*
Scope and conditions
native_context
provider
bedrock_converse*
mode
chat*
full_native_model_id
moonshotai.kimi-k3*
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