Skip to content

Kimi · Model profile

Kimi K2.6

Intelligence, Performance & Price Analysis

Source: Canonical slug: kimi-k2-6 · Canonical model registry at build time

Routing verdict

Frontier research / complex reasoning

Reserve this model for the hardest requests; simpler or repetitive work should stay on cheaper routes.

Kimi K2.6 summary metrics

Intelligence
44 scoreKimi K2.6 scores 44 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 25).
Speed
76.2 output tokens/secKimi K2.6 generates output at 76.2 tokens per second (based on Kimi's API), which is above average compared to other open weight models of similar size (median: 61.7 t/s).
Latency
2.37s TTFTKimi K2.6 has a time to first token (TTFT) of 2.37s (based on Kimi's API), which is somewhat higher than average compared to other open weight models of similar size (median: 2.01s).
Input Price
$0.95 / 1M tokensKimi K2.6 costs $0.95 per 1M input tokens (somewhat higher than average, median: $0.59) and $4.00 per 1M output tokens (at the higher end, median: $2.20), based on Kimi's API.
Output Price
$4.00 / 1M tokensKimi K2.6 costs $0.95 per 1M input tokens (somewhat higher than average, median: $0.59) and $4.00 per 1M output tokens (at the higher end, median: $2.20), based on Kimi's API.
Verbosity
170M Output tokens from Intelligence Index 4 out of 4 units for Verbosity . Comp

Assessment

Routing verdict and task fit

Kimi K2.6 scores 44 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 25). Kimi K2.6 generates output at 76.2 tokens per second (based on Kimi's API), which is above average compared to other open weight models of similar size (median: 61.7 t/s). Kimi K2.6 costs $0.95 per 1M input tokens (somewhat higher than average, median: $0.59) and $4.00 per 1M output tokens (at the higher end, median: $2.20), based on Kimi's API.
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsOptimalIntelligence score 44 places this model among top performers. Suitable for multi-step analysis and agentic loops.
High-volume chat & customer-facingOptimalOutput speed 76.2 tokens/sec is adequate for chat.
Latency-sensitive applicationsOptimalTTFT 2.37s — adequate latency for most interactive use cases.
Cost-sensitive pipelinesViableOutput pricing at $4.00 is premium. Route high-volume simple tasks to cheaper alternatives.

Evidence

Evidence charts

17 published comparison charts for this profile. Kimi K2.6 is marked with a diamond and bold label; each chart has a table view and lists its source and retrieval date.

AA-Omniscience Index

AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct. · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 6.4, row 5 of 20 in the published order.
  1. Claude Fable 5 (with fallback)40.1
  2. Gemini 3.5 Flash22.7
  3. GPT-5.5 (xhigh)20.1
  4. Grok 4.3 (high)18.3
  5. Kimi K2.6 (this model)6.4
  6. Muse Spark4.1
  7. GLM-5.2 (max)4
  8. MiMo-V2.5-Pro3.6
  9. GLM-5.11.9
  10. MiniMax-M31.4
  11. MiniMax-M2.70.683
  12. Nemotron 3 Ultra-0.833
  13. Command A+-4
  14. DeepSeek V4 Pro (max)-10
  15. Kimi K2.7 Code-10.7
  16. DeepSeek V4 Flash (max)-22.9
  17. Nex-N2-Pro-27.6
  18. Qwen3.5 397B A17B-29.8
  19. DeepSeek R1 (Jan)-31.3
  20. GLM-4.7-34.6

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 44.2, row 7 of 20 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.6 (this model)44.2
  8. Muse Spark43.1
  9. MiMo-V2.5-Pro42.2
  10. Kimi K2.7 Code41.9
  11. Nex-N2-Pro41
  12. DeepSeek V4 Flash (max)40.3
  13. GLM-5.140.2
  14. MiniMax-M2.738.1
  15. Nemotron 3 Ultra37.8
  16. Grok 4.3 (high)37.6
  17. GLM-4.733.7
  18. Qwen3.5 397B A17B33.7
  19. Ring-2.6-1T30.6
  20. Step 3.7 Flash29.7

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 44.2, row 7 of 20 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.6 (this model)44.2
  8. Muse Spark43.1
  9. MiMo-V2.5-Pro42.2
  10. Kimi K2.7 Code41.9
  11. Nex-N2-Pro41
  12. DeepSeek V4 Flash (max)40.3
  13. GLM-5.140.2
  14. MiniMax-M2.738.1
  15. Nemotron 3 Ultra37.8
  16. Grok 4.3 (high)37.6
  17. GLM-4.733.7
  18. Qwen3.5 397B A17B33.7
  19. Ring-2.6-1T30.6
  20. Step 3.7 Flash29.7

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open) · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 33.3, row 18 of 20 in the published order.
  1. Nemotron 3 Ultra83.3
  2. MiMo-V2-Flash52.8
  3. DeepSeek V4 Pro (max)50
  4. DeepSeek V4 Flash (max)50
  5. GLM-5.2 (max)44.4
  6. GLM-5.144.4
  7. GLM-4.744.4
  8. GLM-4.644.4
  9. Qwen3 235B A22B 250744.4
  10. gpt-oss-120b (high)38.9
  11. MiMo-V2.5-Pro38.9
  12. Nex-N2-Pro38.9
  13. Qwen3.5 397B A17B38.9
  14. Ring-2.6-1T38.9
  15. Step 3.7 Flash38.9
  16. Command A+38.9
  17. Mistral Large 338.9
  18. Kimi K2.6 (this model)33.3
  19. MiniMax-M333.3
  20. Kimi K2.7 Code27.8

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Intelligence

Artificial Analysis Intelligence Index · Higher is better · Evaluation results measured independently by Artificial Analysis

11 models in the published order. Kimi K2.6: 44.2, row 7 of 11 in the published order.
  1. Claude Fable 5 (with fallback)59.9
  2. GPT-5.5 (xhigh)54.8
  3. GLM-5.2 (max)51.1
  4. Gemini 3.5 Flash50.2
  5. MiniMax-M344.4
  6. DeepSeek V4 Pro (max)44.3
  7. Kimi K2.6 (this model)44.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Output Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 76 t/s, row 15 of 20 in the published order.
  1. Step 3.7 Flash410 t/s
  2. gpt-oss-120b (high)314 t/s
  3. Nemotron 3 Ultra250 t/s
  4. GLM-5.2 (max)218 t/s
  5. Command A+196 t/s
  6. Gemini 3.5 Flash192 t/s
  7. Grok 4.3 (high)164 t/s
  8. Ring-2.6-1T136 t/s
  9. GLM-4.7126 t/s
  10. DeepSeek V4 Flash (max)112 t/s
  11. MiniMax-M399 t/s
  12. Nex-N2-Pro92 t/s
  13. GPT-5.5 (xhigh)88 t/s
  14. GLM-5.180 t/s
  15. Kimi K2.6 (this model)76 t/s
  16. DeepSeek V4 Pro (max)72 t/s
  17. Claude Fable 5 (with fallback)71 t/s
  18. Qwen3 235B A22B 250761 t/s
  19. Mistral Large 353 t/s
  20. MiMo-V2.5-Pro53 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

Speed

Output tokens per second · Higher is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. Kimi K2.6: 76 t/s, row 8 of 10 in the published order.
  1. gpt-oss-120b (high)314 t/s
  2. Nemotron 3 Ultra250 t/s
  3. GLM-5.2 (max)218 t/s
  4. Gemini 3.5 Flash192 t/s
  5. Grok 4.3 (high)164 t/s
  6. MiniMax-M399 t/s
  7. GPT-5.5 (xhigh)88 t/s
  8. Kimi K2.6 (this model)76 t/s
  9. DeepSeek V4 Pro (max)72 t/s
  10. Claude Fable 5 (with fallback)71 t/s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Output speed context
Speed should be read together with latency and cost; fast output does not always mean better user-perceived performance.

End-to-End Response Time

Seconds to output 500 tokens, including reasoning model 'thinking' time · Lower is better · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 6.56 s, row 20 of 20 in the published order.
  1. Step 3.7 Flash1.22 s
  2. gpt-oss-120b (high)1.59 s
  3. Mistral Large 39.43 s
  4. Nemotron 3 Ultra2 s
  5. GLM-5.2 (max)2.3 s
  6. Command A+2.55 s
  7. GLM-4.73.97 s
  8. Ring-2.6-1T3.67 s
  9. Grok 4.3 (high)3.05 s
  10. MiniMax-M35.05 s
  11. Nex-N2-Pro5.42 s
  12. Gemini 3.5 Flash2.6 s
  13. Qwen3 235B A22B 25078.14 s
  14. MiMo-V2.5-Pro9.52 s
  15. GLM-4.610.18 s
  16. GLM-5.16.28 s
  17. DeepSeek V4 Flash (max)4.46 s
  18. Kimi K2.7 Code9.72 s
  19. MiniMax-M2.79.79 s
  20. Kimi K2.6 (this model)6.56 s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 58.43 s, row 20 of 20 in the published order.
  1. Mistral Large 30 s
  2. Step 3.7 Flash4.88 s
  3. gpt-oss-120b (high)6.37 s
  4. Nemotron 3 Ultra9.09 s
  5. Command A+10.18 s
  6. GLM-5.2 (max)9.19 s
  7. GLM-4.715.87 s
  8. Ring-2.6-1T14.68 s
  9. Grok 4.3 (high)0 s
  10. MiniMax-M320.19 s
  11. Nex-N2-Pro21.66 s
  12. Gemini 3.5 Flash0 s
  13. Qwen3 235B A22B 250732.55 s
  14. MiMo-V2.5-Pro38.08 s
  15. GLM-4.640.7 s
  16. Kimi K2.7 Code43.34 s
  17. GLM-5.147.62 s
  18. MiniMax-M2.748.19 s
  19. DeepSeek V4 Flash (max)50.09 s
  20. Kimi K2.6 (this model)58.43 s

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Latency considerations
Latency includes thinking / reasoning time when applicable. Lower is better for interactive use cases.

Context Window

Context window: tokens limit · Higher is better · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 256k, row 16 of 20 in the published order.
  1. Gemini 3.5 Flash1M
  2. Claude Fable 5 (with fallback)1M
  3. GLM-5.2 (max)1M
  4. DeepSeek V4 Pro (max)1M
  5. Grok 4.3 (high)1M
  6. MiniMax-M31M
  7. MiMo-V2.5-Pro1M
  8. DeepSeek V4 Flash (max)1M
  9. GPT-5.5 (xhigh)922k
  10. Nemotron 3 Ultra262.1k
  11. Muse Spark262.1k
  12. Qwen3.5 397B A17B262.1k
  13. Ring-2.6-1T262.1k
  14. Step 3.7 Flash262.1k
  15. Nex-N2-Pro262k
  16. Kimi K2.6 (this model)256k
  17. Kimi K2.7 Code256k
  18. MiMo-V2-Flash256k
  19. Qwen3 235B A22B 2507256k
  20. Mistral Large 3256k

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Context window sizing
Larger context windows enable complex multi-turn tasks but may increase latency and cost for long inputs.

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 0.0216, row 15 of 20 in the published order.
  1. MiMo-V2.5-Pro0.00466
  2. DeepSeek V4 Pro (max)0.0063
  3. Mistral Large 30.00455
  4. gpt-oss-120b (high)0.00454
  5. MiniMax-M30.0139
  6. Grok 4.3 (high)0.0101
  7. Kimi K2.7 Code0.0203
  8. GLM-5.10.0426
  9. Nemotron 3 Ultra0.0209
  10. DeepSeek R1 (Jan)0.0153
  11. GLM-4.60.0144
  12. GLM-4.70.0157
  13. Qwen3.5 397B A17B0.022
  14. Ring-2.6-1T0.0239
  15. Kimi K2.6 (this model)0.0216
  16. MiMo-V2-Flash0.00812
  17. GLM-5.2 (max)0.0268
  18. Gemini 3.5 Flash0.0884
  19. GPT-5.5 (xhigh)0.107
  20. Claude Fable 5 (with fallback)0.367

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better · Evaluation results measured independently by Artificial Analysis

10 models in the published order. Kimi K2.6: $0.346, row 6 of 10 in the published order.
  1. DeepSeek V4 Pro (max)$0.045
  2. gpt-oss-120b (high)$0.061
  3. MiniMax-M3$0.125
  4. Grok 4.3 (high)$0.139
  5. Nemotron 3 Ultra$0.244
  6. Kimi K2.6 (this model)$0.346
  7. GLM-5.2 (max)$0.370
  8. Gemini 3.5 Flash$0.586
  9. GPT-5.5 (xhigh)$0.863
  10. Claude Fable 5 (with fallback)$2.75

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6: 25, row 17 of 20 in the published order.
  1. Mistral Large 38.4
  2. gpt-oss-120b (high)3.5
  3. MiMo-V2.5-Pro6.7
  4. DeepSeek V4 Pro (max)7.2
  5. MiniMax-M314.5
  6. Grok 4.3 (high)12.5
  7. GLM-4.615.6
  8. MiMo-V2-Flash11.4
  9. DeepSeek R1 (Jan)22
  10. Nemotron 3 Ultra33.5
  11. Ring-2.6-1T23.5
  12. Kimi K2.7 Code24.3
  13. Qwen3.5 397B A17B24.9
  14. GLM-4.716.6
  15. GLM-5.140.4
  16. GLM-5.2 (max)32.9
  17. Kimi K2.6 (this model)25
  18. Gemini 3.5 Flash114
  19. GPT-5.5 (xhigh)131.8
  20. Claude Fable 5 (with fallback)507.6

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.
Cost pressure
Ethen Gateway should route away from this model when the task is simple, repetitive, or likely to generate many output tokens.

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens) · Evaluation results measured independently by Artificial Analysis

20 models; segments: Input, Output, Cache Hit. Kimi K2.6: $5.11, row 17 of 20 in the published order.
  • Input
  • Output
  • Cache Hit
  1. MiMo-V2-Flash$0.400
  2. DeepSeek V4 Flash (max)$0.423
  3. gpt-oss-120b (high)$0.900
  4. DeepSeek V4 Pro (max)$1.31
  5. MiMo-V2.5-Pro$1.31
  6. Step 3.7 Flash$1.39
  7. MiniMax-M3$1.56
  8. MiniMax-M2.7$1.56
  9. Mistral Large 3$2.00
  10. GLM-4.6$2.75
  11. Ring-2.6-1T$2.80
  12. GLM-4.7$3.33
  13. Nex-N2-Pro$3.25
  14. Nemotron 3 Ultra$3.60
  15. Grok 4.3 (high)$3.95
  16. Qwen3.5 397B A17B$4.20
  17. Kimi K2.6 (this model)$5.11
  18. Kimi K2.7 Code$5.11
  19. GLM-5.2 (max)$6.06
  20. GLM-5.1$6.06

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Chart interpretation
Each bar is one model measured the same way, so this model can be read against the others. The chart lists its source and retrieval date.

Time per Intelligence Index Task

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6 is not marked in this published set of 20 models.
  1. Mistral Large 30.924
  2. Nemotron 3 Ultra1.4
  3. Grok 4.3 (high)1.6
  4. Command A+1.7
  5. Step 3.7 Flash1.9
  6. gpt-oss-120b (high)1.9
  7. Gemini 3.5 Flash2.2
  8. GLM-4.72.5
  9. GLM-5.2 (max)2.8
  10. Ring-2.6-1T2.9
  11. GPT-5.5 (xhigh)3
  12. MiniMax-M33.9
  13. Qwen3 235B A22B 25074.4
  14. Claude Fable 5 (with fallback)4.8
  15. Nex-N2-Pro4.8
  16. GLM-5.14.9
  17. Kimi K2.7 Code5
  18. Qwen3.5 397B A17B5.1
  19. MiniMax-M2.75.5
  20. GLM-4.65.9

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference · Evaluation results measured independently by Artificial Analysis

20 models; segments: Passive Params, Active Params. Kimi K2.6: 1,000B, row 3 of 20 in the published order.
  • Passive Params
  • Active Params
  1. DeepSeek V4 Pro (max)1,600B
  2. MiMo-V2.5-Pro1,023B
  3. Kimi K2.6 (this model)1,000B
  4. Kimi K2.7 Code1,000B
  5. Ring-2.6-1T1,000B
  6. GLM-5.2 (max)753B
  7. GLM-5.1744B
  8. DeepSeek R1 (Jan)685B
  9. Mistral Large 3675B
  10. Nemotron 3 Ultra550B
  11. MiniMax-M3428B
  12. Nex-N2-Pro397B
  13. Qwen3.5 397B A17B397B
  14. GLM-4.7357B
  15. GLM-4.6357B
  16. MiMo-V2-Flash309B
  17. DeepSeek V4 Flash (max)284B
  18. Qwen3 235B A22B 2507235B
  19. MiniMax-M2.7230B
  20. Command A+218B

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Architecture scale
Larger models generally perform better but cost more to serve. Active parameter count matters for inference efficiency.

Output Tokens per Intelligence Index Task

Weighted average number of output tokens used to run one task in the Artificial Analysis Intelligence Index · Evaluation results measured independently by Artificial Analysis

20 models in the published order. Kimi K2.6 is not marked in this published set of 20 models.
  1. Mistral Large 33,031.1
  2. Muse Spark2,857.2
  3. Grok 4.3 (high)4,032
  4. DeepSeek R1 (Jan)3,259.2
  5. GPT-5.5 (xhigh)3,572.5
  6. Qwen3.5 397B A17B6,116.6
  7. Qwen3 235B A22B 25071,604.5
  8. Kimi K2.7 Code5,065.7
  9. MiniMax-M2.77,874.3
  10. GLM-4.66,527.7
  11. MiMo-V2.5-Pro5,359.6
  12. Command A+6,202
  13. Nemotron 3 Ultra7,796.7
  14. GLM-4.77,151.4
  15. MiniMax-M311,623.4
  16. GLM-5.19,684.8
  17. MiMo-V2-Flash27,074.7
  18. Ring-2.6-1T9,565.1
  19. Gemini 3.5 Flash9,819.4
  20. Nex-N2-Pro6,165.3

Published benchmark dataset · Retrieved

Method and source
Methodology
Independent test run by Artificial Analysis on dedicated hardware.
Benchmark interpretation
Higher values indicate stronger overall capability. Bars show how models compare on composite intelligence metrics.

Specifications

Architecture and limits

Kimi K2.6 technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningYesfaq
Input modalitiesKimi K2.6 supports text, image, and video input.faq
Output modalitiesKimi K2.6 supports text output.faq
Context window260k tokensfaq
Open weights / sourceYes, Kimi K2.6 is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersKimi K2.6 has 1 trillion parameters (32 billion active).faq
Active parametersKimi K2.6 is a Mixture of Experts (MoE) model with 1 trillion total parameters, but only 32 billion active parameters are used during inference.faq
LicenseKimi K2.6 is released under the Modified MIT license. This license allows commercial use.faq
API availabilityYes, Kimi K2.6 is available via API through 16 providers.faq

Method

Methodology & Provenance

This page is rendered from the normalized profile and page JSON for Kimi K2.6.

Benchmark values are preserved as normalized; only layout, disclosure ordering, and typography are adjusted for readability.

Questions

Model FAQ and technical disclosures

Kimi K2.6 was released on April 20, 2026.

Kimi K2.6 was created by Kimi.

Kimi K2.6 scores 44 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 25).

Kimi K2.6 generates output at 76.2 tokens per second (based on Kimi's API), which is above average compared to other open weight models of similar size (median: 61.7 t/s).

Kimi K2.6 has a time to first token (TTFT) of 2.37s (based on Kimi's API), which is somewhat higher than average compared to other open weight models of similar size (median: 2.01s).

Kimi K2.6 costs $0.95 per 1M input tokens (somewhat higher than average, median: $0.59) and $4.00 per 1M output tokens (at the higher end, median: $2.20), based on Kimi's API.

Kimi K2.6 costs $0.95 per 1M input tokens and $4.00 per 1M output tokens (based on Kimi's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.70 per 1M tokens. Pricing may vary by provider.

When evaluated on the Intelligence Index, Kimi K2.6 generated 170M output tokens, which is at the higher end compared to other open weight models of similar size (median: 92M).

Yes, Kimi K2.6 is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.

Kimi K2.6 supports text, image, and video input.

Kimi K2.6 supports text output.

Yes, Kimi K2.6 supports image input and can analyze, describe, and answer questions about images.

Yes, Kimi K2.6 is multimodal. It can process text, image, and video input and generate text output.

Kimi K2.6 has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Kimi K2.6 is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Kimi K2.6 has 1 trillion parameters (32 billion active).

Kimi K2.6 is a Mixture of Experts (MoE) model with 1 trillion total parameters, but only 32 billion active parameters are used during inference.

Kimi K2.6 is released under the Modified MIT license. This license allows commercial use.

Kimi K2.6 achieves a score of 44 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Kimi K2.6 is available via API through 16 providers.

Kimi K2.6 is available through 16 API providers.

Models

  • 01KimiOther Kimi models in the catalog.
  • 02LeaderboardsCertified metric ranks, when published.