AI Index

Models tracked
327
Labs
49
Benchmarked
653
Released, 12 months
202
USD per 1M tokens, blended 3:1
Artificial Analysis Intelligence Index
653 models
  1. 1

    Claude Fable 5.1

    Adaptive Reasoning, Max Effort, Default Fallback
    Anthropic Reasoning
    Intelligence
    53.4
    Coding
    81.6
    Speed
    73 t/s
    Price
    $20/1M
    In $10Out $50TTFT 151.53sCtx 1M
  2. 2

    Claude Fable 5.1

    Adaptive Reasoning, Xhigh Effort, Default Fallback
    Anthropic Reasoning
    Intelligence
    53.2
    Coding
    80.7
    Speed
    57 t/s
    Price
    $20/1M
    In $10Out $50TTFT 30.35sCtx 1M
  3. 3

    GPT-6 Astra

    max
    OpenAI Reasoning
    Intelligence
    52.8
    Coding
    76.9
    Speed
    59 t/s
    Price
    $20/1M
    In $10Out $50TTFT 214.47sCtx 1.05M
  4. 4

    GPT-6 Astra

    xhigh
    OpenAI Reasoning
    Intelligence
    52.5
    Coding
    75.9
    Speed
    58 t/s
    Price
    $20/1M
    In $10Out $50TTFT 70.49sCtx 1.05M
  5. 5

    Claude Fable 5.1

    Adaptive Reasoning, High Effort, Default Fallback
    Anthropic Reasoning
    Intelligence
    51.2
    Coding
    79.1
    Speed
    55 t/s
    Price
    $20/1M
    In $10Out $50TTFT 8.64sCtx 1M
  6. 6

    GPT-6 Astra

    high
    OpenAI Reasoning
    Intelligence
    51.0
    Coding
    77.1
    Speed
    54 t/s
    Price
    $20/1M
    In $10Out $50TTFT 26.87sCtx 1.05M
  7. 7

    Claude Opus 5

    Adaptive Reasoning, Max Effort
    Anthropic Reasoning
    Intelligence
    50.7
    Coding
    78.0
    Speed
    54 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 25.73sCtx 1M
  8. 8

    Claude Fable 5

    Adaptive Reasoning, Max Effort, Opus 4.8 Fallback
    Anthropic Reasoning
    Intelligence
    49.7
    Coding
    76.5
    Speed
    64 t/s
    Price
    $20/1M
    In $10Out $50TTFT 44.13sCtx 1M
  9. 9

    Claude Opus 5

    Adaptive Reasoning, Xhigh Effort
    Anthropic Reasoning
    Intelligence
    49.7
    Coding
    77.0
    Speed
    53 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 17.59sCtx 1M
  10. 10

    GPT-6 Astra

    medium
    OpenAI Reasoning
    Intelligence
    49.7
    Coding
    76.7
    Speed
    53 t/s
    Price
    $20/1M
    In $10Out $50TTFT 4.05sCtx 1.05M
  11. 11

    Claude Fable 5.1

    Adaptive Reasoning, Medium Effort, Default Fallback
    Anthropic Reasoning
    Intelligence
    49.1
    Coding
    77.1
    Speed
    51 t/s
    Price
    $20/1M
    In $10Out $50TTFT 5.91sCtx 1M
  12. 12

    Claude Opus 5

    Adaptive Reasoning, High Effort
    Anthropic Reasoning
    Intelligence
    48.2
    Coding
    76.5
    Speed
    50 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 7.84sCtx 1M
  13. 13

    Muse Spark 1.3

    max
    Meta Reasoning
    Intelligence
    48.2
    Coding
    75.8
    Speed
    234 t/s
    Price
    $2.00/1M
    In $1.25Out $4.25TTFT 23.42sCtx 1.05M
  14. 14

    GPT-5.6 Sol

    max
    OpenAI Reasoning
    Intelligence
    47.1
    Coding
    77.4
    Speed
    65 t/s
    Price
    $8.00/1M
    In $4.00Out $20TTFT 71.18sCtx 1.05M
  15. 15

    Claude Fable 5.1

    Adaptive Reasoning, Low Effort, Default Fallback
    Anthropic Reasoning
    Intelligence
    47.0
    Coding
    75.2
    Speed
    50 t/s
    Price
    $20/1M
    In $10Out $50TTFT 4.02sCtx 1M
  16. 16

    GPT-6 Astra

    low
    OpenAI Reasoning
    Intelligence
    46.0
    Coding
    75.7
    Speed
    54 t/s
    Price
    $20/1M
    In $10Out $50TTFT 1.33sCtx 1.05M
  17. 17

    Qwen3.8 Max

    0902
    Alibaba Reasoning
    Intelligence
    45.4
    Coding
    76.2
    Speed
    40 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 1.66sCtx 1M
  18. 18

    Muse Spark 1.3

    xhigh
    Meta Reasoning
    Intelligence
    45.2
    Coding
    76.5
    Speed
    223 t/s
    Price
    $2.00/1M
    In $1.25Out $4.25TTFT 20.72sCtx 1.05M
  19. 19

    Claude Opus 5

    Adaptive Reasoning, Medium Effort
    Anthropic Reasoning
    Intelligence
    45.1
    Coding
    74.3
    Speed
    49 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 3.40sCtx 1M
  20. 20

    GLM-5.3

    max
    Z.ai Reasoning Open weights
    Intelligence
    44.9
    Coding
    74.8
    Speed
    72 t/s
    Price
    $2.15/1M
    In $1.40Out $4.40TTFT 2.65sCtx 1.31M
  21. 21

    Grok 4.6

    high
    xAI Reasoning
    Intelligence
    44.4
    Coding
    76.8
    Speed
    57 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 31.81sCtx 500K
  22. 22

    Grok 4.6

    xhigh
    xAI Reasoning
    Intelligence
    44.3
    Coding
    75.9
    Speed
    61 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 35.96sCtx 500K
  23. 23

    GPT-5.6 Sol

    xhigh
    OpenAI Reasoning
    Intelligence
    44.1
    Coding
    78.3
    Speed
    60 t/s
    Price
    $8.00/1M
    In $4.00Out $20TTFT 27.34sCtx 1.05M
  24. 24

    Kimi K3

    max
    Kimi Reasoning Open weights
    Intelligence
    43.8
    Coding
    76.2
    Speed
    39 t/s
    Price
    $6.00/1M
    In $3.00Out $15TTFT 3.18s
  25. 25

    Step 5 Preview

    StepFun Open weights
    Intelligence
    43.6
    Coding
    —
    Speed
    93 t/s
    Price
    $1.43/1M
    In $1.00Out $2.70TTFT 1.80s
  26. 26

    Grok 4.6

    medium
    xAI Reasoning
    Intelligence
    43.0
    Coding
    74.4
    Speed
    59 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 23.21sCtx 500K
  27. 27

    GPT-5.6 Sol

    high
    OpenAI Reasoning
    Intelligence
    42.5
    Coding
    77.2
    Speed
    56 t/s
    Price
    $8.00/1M
    In $4.00Out $20TTFT 10.51sCtx 1.05M
  28. 28

    GPT-5.6 Terra

    max
    OpenAI Reasoning
    Intelligence
    42.3
    Coding
    76.7
    Speed
    88 t/s
    Price
    $4.50/1M
    In $2.00Out $12TTFT 123.55sCtx 1.05M
  29. 29

    Claude Opus 4.8

    Adaptive Reasoning, Max Effort
    Anthropic Reasoning
    Intelligence
    42.0
    Coding
    74.3
    Speed
    0 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 0.00sCtx 1M
  30. 30

    GLM 5.3 Flash

    Z.ai Reasoning Open weights
    Intelligence
    41.9
    Coding
    71.5
    Speed
    94 t/s
    Price
    $0.24/1M
    In $0.15Out $0.5TTFT 2.40sCtx 1.31M
  31. 31

    Gemini 3.8 Flash

    high
    Google Reasoning
    Intelligence
    41.2
    Coding
    76.3
    Speed
    314 t/s
    Price
    $1.50/1M
    In $0.75Out $3.75TTFT 10.73sCtx 1.05M
  32. 32

    Claude Opus 4.7

    Adaptive Reasoning, Max Effort
    Anthropic Reasoning
    Intelligence
    40.7
    Coding
    73.6
    Speed
    0 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 0.00sCtx 1M
  33. 33

    Qwen3.8 Max

    Alibaba Reasoning
    Intelligence
    40.3
    Coding
    71.8
    Speed
    45 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 1.55sCtx 1M
  34. 34

    Gemini 3.8 Flash

    medium
    Google Reasoning
    Intelligence
    40.0
    Coding
    74.1
    Speed
    0 t/s
    Price
    $1.50/1M
    In $0.75Out $3.75TTFT 0.00sCtx 1.05M
  35. 35

    Qwen3.8 2.4T A95B

    Alibaba Reasoning Open weights
    Intelligence
    40.0
    Coding
    71.9
    Speed
    39 t/s
    Price
    $3.00/1M
    In $2.00Out $6.00TTFT 1.86sCtx 1.05M
  36. 36

    Qwen3.8-Flash-Next

    Alibaba
    Intelligence
    39.9
    Coding
    73.1
    Speed
    52 t/s
    Price
    $0.23/1M
    In $0.15Out $0.47TTFT 1.42s
  37. 37

    Claude Opus 5

    Adaptive Reasoning, Low Effort
    Anthropic Reasoning
    Intelligence
    39.8
    Coding
    66.9
    Speed
    53 t/s
    Price
    $10/1M
    In $5.00Out $25TTFT 2.29sCtx 1M
  38. 38

    Muse Spark 1.2

    xhigh
    Meta Reasoning
    Intelligence
    39.8
    Coding
    72.2
    Speed
    225 t/s
    Price
    $2.00/1M
    In $1.25Out $4.25TTFT 10.24sCtx 1.05M
  39. 39

    Gemini 3.7 Flash

    medium
    Google Reasoning
    Intelligence
    39.6
    Coding
    71.5
    Speed
    345 t/s
    Price
    $1.50/1M
    In $0.75Out $3.75TTFT 5.27sCtx 1.05M
  40. 40

    DeepSeek V4.1 Flash

    Reasoning, Max Effort
    DeepSeek Reasoning Open weights
    Intelligence
    39.5
    Coding
    —
    Speed
    223 t/s
    Price
    $0.53/1M
    In $0.3Out $1.20TTFT 0.82sCtx 1.05M

Benchmarks, speed and prices: Artificial Analysis · capabilities and catalogue: OpenRouter

  • Highest intelligence Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)Anthropic53.4 Intelligence Index
  • Fastest output Celeris-1Celeris2,027 t/s tokens per second
  • Lowest latency Claude Opus 4.8 (Adaptive Reasoning, Max Effort)Anthropic0.00 s time to first token
  • Best value Step 5 PreviewStepFun$1.43/1M cheapest within 20% of the top score
  • Largest context Grok 4.20xAI2M tokens
  • Newest frontier model GPT-6 AstraOpenAI4 Sep 2026 flagship-class release

Intelligence Index

Composite of reasoning, knowledge, maths and coding evaluations · higher is better

Artificial Analysis
Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback): 5353Claude Fable 5.1 (Adaptiv…Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback): 5353Claude Fable 5.1 (Adaptiv…GPT-6 Astra (max): 5353GPT-6 Astra (max)GPT-6 Astra (xhigh): 5353GPT-6 Astra (xhigh)Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback): 5151Claude Fable 5.1 (Adaptiv…GPT-6 Astra (high): 5151GPT-6 Astra (high)Claude Opus 5 (Adaptive Reasoning, Max Effort): 5151Claude Opus 5 (Adaptive R…Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback): 5050Claude Fable 5 (Adaptive …Claude Opus 5 (Adaptive Reasoning, Xhigh Effort): 5050Claude Opus 5 (Adaptive R…GPT-6 Astra (medium): 5050GPT-6 Astra (medium)Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback): 4949Claude Fable 5.1 (Adaptiv…Claude Opus 5 (Adaptive Reasoning, High Effort): 4848Claude Opus 5 (Adaptive R…Muse Spark 1.3 (max): 4848Muse Spark 1.3 (max)GPT-5.6 Sol (max): 4747GPT-5.6 Sol (max)Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback): 4747Claude Fable 5.1 (Adaptiv…GPT-6 Astra (low): 4646GPT-6 Astra (low)Qwen3.8 Max (0902): 4545Qwen3.8 Max (0902)Muse Spark 1.3 (xhigh): 4545Muse Spark 1.3 (xhigh)Claude Opus 5 (Adaptive Reasoning, Medium Effort): 4545Claude Opus 5 (Adaptive R…GLM-5.3 (max): 4545GLM-5.3 (max)Grok 4.6 (high): 4444Grok 4.6 (high)Grok 4.6 (xhigh): 4444Grok 4.6 (xhigh)GPT-5.6 Sol (xhigh): 4444GPT-5.6 Sol (xhigh)Kimi K3 (max): 4444Kimi K3 (max)
  • Anthropic
  • OpenAI
  • Meta
  • Alibaba
  • Z.ai
  • xAI
  • Kimi

Intelligence vs. price

Blended price per 1M tokens (3:1 input:output) against Intelligence Index · top-left is best · log price axis

Artificial Analysis
$0.01$0.1$1$10$100$1000 0102030405060 Price, USD per 1M tokens Intelligence Index Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback): 53.4 · $20.00/1MClaude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback): 53.2 · $20.00/1MGPT-6 Astra (max): 52.8 · $20.00/1MGPT-6 Astra (xhigh): 52.5 · $20.00/1MClaude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback): 51.2 · $20.00/1MGPT-6 Astra (high): 51.0 · $20.00/1MClaude Opus 5 (Adaptive Reasoning, Max Effort): 50.7 · $10.00/1MClaude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback): 49.7 · $20.00/1MClaude Opus 5 (Adaptive Reasoning, Xhigh Effort): 49.7 · $10.00/1MGPT-6 Astra (medium): 49.7 · $20.00/1MClaude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback): 49.1 · $20.00/1MClaude Opus 5 (Adaptive Reasoning, High Effort): 48.2 · $10.00/1MMuse Spark 1.3 (max): 48.2 · $2.00/1MGPT-5.6 Sol (max): 47.1 · $8.00/1MClaude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback): 47.0 · $20.00/1MGPT-6 Astra (low): 46.0 · $20.00/1MQwen3.8 Max (0902): 45.4 · $3.00/1MMuse Spark 1.3 (xhigh): 45.2 · $2.00/1MClaude Opus 5 (Adaptive Reasoning, Medium Effort): 45.1 · $10.00/1MGLM-5.3 (max): 44.9 · $2.15/1MGrok 4.6 (high): 44.4 · $3.00/1MGrok 4.6 (xhigh): 44.3 · $3.00/1MGPT-5.6 Sol (xhigh): 44.1 · $8.00/1MKimi K3 (max): 43.8 · $6.00/1MStep 5 Preview: 43.6 · $1.43/1MGrok 4.6 (medium): 43.0 · $3.00/1MGPT-5.6 Sol (high): 42.5 · $8.00/1MGPT-5.6 Terra (max): 42.3 · $4.50/1MClaude Opus 4.8 (Adaptive Reasoning, Max Effort): 42.0 · $10.00/1MGLM 5.3 Flash: 41.9 · $0.24/1MGemini 3.8 Flash (high): 41.2 · $1.50/1MClaude Opus 4.7 (Adaptive Reasoning, Max Effort): 40.7 · $10.00/1MQwen3.8 Max: 40.3 · $3.00/1MGemini 3.8 Flash (medium): 40.0 · $1.50/1MQwen3.8 2.4T A95B: 40.0 · $3.00/1MQwen3.8-Flash-Next: 39.9 · $0.23/1MClaude Opus 5 (Adaptive Reasoning, Low Effort): 39.8 · $10.00/1MMuse Spark 1.2 (xhigh): 39.8 · $2.00/1MGemini 3.7 Flash (medium): 39.6 · $1.50/1MDeepSeek V4.1 Flash (Reasoning, Max Effort): 39.5 · $0.53/1MGPT-5.6 Sol (medium): 39.5 · $8.00/1MGemini 3.7 Flash (high): 39.4 · $1.50/1MGrok 4.5 (high): 39.1 · $3.00/1MGPT-5.4 (xhigh): 39.0 · $5.63/1MGPT-5.5 (xhigh): 38.6 · $11.25/1MClaude Sonnet 5 (Adaptive Reasoning, Max Effort): 38.4 · $4.00/1MGPT-5.6 Terra (xhigh): 38.2 · $4.50/1MGPT-5.6 Luna (max): 37.5 · $0.45/1MGPT-5.5 (high): 37.3 · $11.25/1MGemini 3.7 Flash (low): 36.9 · $1.50/1MDeepSeek V4 Pro 0813 (Reasoning, Max Effort): 36.3 · $1.98/1MAgnes 3.0 Flash: 35.5 · $0.08/1MGrok 4.6 (low): 35.4 · $3.00/1MAgnes 2.5 Pro Beta: 35.2 · $0.15/1MDeepSeek V4 Flash Vision (Reasoning, Max Effort): 35.0 · $0.66/1MGPT-5.6 Luna (xhigh): 34.8 · $0.45/1MClaude Sonnet 5 (Adaptive Reasoning, Xhigh Effort): 34.7 · $4.00/1MDeepSeek V4 Flash 0731 (Reasoning, Max Effort): 34.5 · $0.66/1MGPT-5.6 Terra (high): 34.5 · $4.50/1MGemini 3.6 Flash (high): 34.3 · $1.50/1MMuse Spark 1.1 (xhigh): 34.3 · $2.00/1MGPT-5.5 (medium): 34.2 · $11.25/1MGLM-5.2 (max): 34.0 · $2.15/1MQwen3.8 27B (xhigh): 33.9 · $1.13/1MGemini 3.8 Flash (low): 33.8 · $1.50/1MGPT-5.6 Sol (low): 33.8 · $8.00/1MGemini 3.5 Flash (medium): 33.6 · $3.38/1MGemini 3.5 Flash (high): 33.0 · $3.38/1MGPT-5.3 Codex (xhigh): 32.5 · $4.81/1MGPT-5.6 Luna (high): 32.4 · $0.45/1MClaude Sonnet 5 (Adaptive Reasoning, High Effort): 32.0 · $4.00/1MClaude Opus 4.6 (Adaptive Reasoning, Max Effort): 31.9 · $10.00/1MClaude Opus 4.7 (Non-reasoning, High Effort): 30.9 · $10.00/1MDeepSeek V4 Pro 0424 (Reasoning, Max Effort): 30.9 · $0.54/1MGPT-5.5 (low): 30.7 · $11.25/1MClaude Sonnet 4.6 (Adaptive Reasoning, Max Effort): 30.5 · $6.00/1MKimi K3 (low): 30.5 · $6.00/1MApodex 1.1: 30.4 · $0.98/1MGemini 3.1 Pro Preview: 30.4 · $4.50/1MGPT-5.2 (xhigh): 30.4 · $4.81/1MGPT-5.6 Terra (medium): 30.4 · $4.50/1MDeepSeek V4 Pro 0424 (Reasoning, High Effort): 30.1 · $0.54/1MQwen3.7 Max: 29.9 · $3.75/1MMiniMax-M3: 29.6 · $0.53/1MClaude Opus 4.5 (Reasoning): 29.1 · $10.00/1MClaude Sonnet 5 (Non-reasoning, High Effort): 28.9 · $4.00/1MGPT-5.2 Codex (xhigh): 28.5 · $4.81/1MClaude Sonnet 5 (Adaptive Reasoning, Medium Effort): 28.4 · $4.00/1MQwen3.6 Max Preview: 28.4 · $2.93/1MGPT-5.6 Sol (Non-reasoning): 28.3 · $8.00/1MSolar Pro 4: 28.2 · $0.53/1MGemini 3 Pro Preview (high): 28.0 · $4.50/1MGLM-5 (Reasoning): 27.9 · $1.55/1MGPT-5.6 Terra (low): 27.9 · $4.50/1MAgnes 2.5 Pro Alpha: 27.8 · $0.56/1MQwen3.8 27B (medium): 27.8 · $1.13/1MGPT-5.4 (low): 27.6 · $5.63/1MKimi K2.6: 27.5 · $1.71/1MGrok Build 0.1 0616: 27.2 · $1.25/1MQuasar 438B (max, based on GLM-5.2): 27.1 · $0.90/1MQwen3.6 Plus: 27.0 · $1.13/1MGPT-5.5 Instant (June 2026): 26.8 · $11.25/1MGPT-5.2 (medium): 26.5 · $4.81/1MQwen3.8 27B (low): 26.5 · $1.13/1MClaude Opus 4.6 (Non-reasoning, High Effort): 26.4 · $10.00/1MGLM-5.1 (Reasoning): 26.4 · $2.00/1MMiMo-V2.5-Pro: 26.4 · $0.54/1MGemini 3 Flash Preview (Reasoning): 26.3 · $1.13/1MKimi K2.7 Code: 26.3 · $1.71/1MInkling Small: 26.1 · $0.53/1MHy3: 25.8 · $0.25/1MQwen3.7 Plus: 25.8 · $0.70/1MGrok 4.20 0309 v2 (Reasoning): 25.7 · $1.56/1MGPT-5.6 Luna (medium): 25.5 · $0.45/1MInkling (xhigh): 25.5 · $1.76/1MGrok 4.3 (high): 25.4 · $1.56/1MGrok 4.20 0309 (Reasoning): 25.2 · $3.00/1MGPT-5 Codex (high): 24.9 · $3.44/1MDeepSeek V4 Flash 0420 (Reasoning, High Effort): 24.8 · $0.17/1MGrok 4.3 (medium): 24.8 · $1.56/1MClaude Sonnet 4.6 (Non-reasoning, High Effort): 24.7 · $6.00/1MClaude Sonnet 5 (Adaptive Reasoning, Low Effort): 24.7 · $4.00/1MGPT-5.1 (high): 24.7 · $3.44/1MDeepSeek V4 Flash 0420 (Reasoning, Max Effort): 24.6 · $0.17/1MGPT-5.4 mini (xhigh): 24.6 · $1.69/1MGrok 4.3 (low): 24.3 · $1.56/1MGLM-5.1 (Non-reasoning): 24.2 · $2.14/1MGemini 3.5 Flash (minimal): 23.8 · $3.38/1MClaude Opus 4.5 (Non-reasoning): 23.7 · $10.00/1MGPT-5.1 Codex (high): 23.7 · $3.44/1MKimi K2.6 (Non-reasoning): 23.6 · $1.71/1MKimi K2.5 (Reasoning): 23.5 · $1.14/1MNemotron 3 Ultra 550B A55B (Reasoning): 23.4 · $1.05/1MClaude Sonnet 4.6 (Non-reasoning, Low Effort): 23.3 · $6.00/1MGPT-5.5 (Non-reasoning): 23.2 · $11.25/1MMiniMax-M2.7: 23.2 · $0.53/1MGPT-5 (high): 23.0 · $3.44/1MGPT-5 (medium): 22.9 · $3.44/1MQwen3.5 27B (Reasoning): 22.9 · $0.83/1MClaude 4.1 Opus (Reasoning): 22.8 · $30.00/1MMiniMax-M2.5: 22.8 · $0.53/1MGemini 3.5 Flash-Lite: 22.7 · $0.85/1MGPT-5.5 Instant (May 2026): 22.7 · $11.25/1MHy3-preview (Reasoning): 22.7 · $0.10/1MGrok 4: 22.5 · $6.00/1MGLM-5.2 (Non-reasoning): 22.4 · $2.15/1MQwen3.8 27B (Non-reasoning): 22.4 · $1.13/1MGemini 3 Pro Preview (low): 22.3 · $4.50/1MMiMo-V2.5: 22.3 · $0.18/1MGLM-4.7 (Reasoning): 22.2 · $1.00/1MKimi K2 Thinking: 22.0 · $1.08/1Mo3-pro: 21.9 · $35.00/1MQwen3.6 27B (Reasoning): 21.9 · $1.35/1MGLM-5 (Non-reasoning): 21.8 · $1.55/1MKAT Coder Pro V2: 21.7 · $0.53/1MDeepSeek V3.2 (Reasoning): 21.5 · $0.32/1MGPT-5.6 Luna (low): 21.5 · $0.45/1MQwen3.5 397B A17B (Non-reasoning): 21.4 · $1.35/1MClaude 4.5 Sonnet (Reasoning): 21.2 · $6.00/1MGPT-5.4 nano (xhigh): 21.2 · $0.46/1MGPT-5.6 Terra (Non-reasoning): 21.2 · $4.50/1MMiniMax-M2.1: 20.9 · $0.53/1MDeepSeek V4 Pro 0424 (Non-reasoning): 20.8 · $0.54/1MGPT-5 (low): 20.8 · $3.44/1MMiMo-V2-Flash (Reasoning): 20.8 · $0.15/1MClaude 4 Opus (Reasoning): 20.6 · $30.00/1MGPT-5 mini (medium): 20.6 · $0.69/1MLing 3.0 Flash: 20.6 · $0.11/1MGPT-5.1 Codex mini (high): 20.4 · $0.69/1MQwen3.5 Omni Plus: 20.4 · $1.50/1Mo3: 20.2 · $3.50/1MGPT-5.4 nano (medium): 20.0 · $0.46/1MQwen3.6 27B (Non-reasoning): 19.8 · $1.35/1MGPT-5.4 mini (medium): 19.7 · $1.69/1MLongCat 2.0: 19.7 · $0.53/1MStep 3.7 Flash: 19.5 · $0.44/1MKimi K2.5 (Non-reasoning): 19.4 · $1.20/1MQwen3.5 27B (Non-reasoning): 19.4 · $0.83/1MClaude 4.5 Sonnet (Non-reasoning): 19.3 · $6.00/1MQwen3.5 35B A3B (Reasoning): 19.3 · $0.69/1MQwen3.5 397B A17B (Reasoning): 19.1 · $1.35/1MDeepSeek V4 Flash 0420 (Non-reasoning): 18.9 · $0.12/1MQwen3.6 35B A3B (Reasoning): 18.8 · $0.84/1MClaude 4.1 Opus (Non-reasoning): 18.6 · $30.00/1MMiniMax-M2: 18.6 · $0.53/1MGLM-4.6 (Reasoning): 18.5 · $0.96/1MMiMo-V2.5-Pro (Non-reasoning): 18.3 · $0.54/1MGPT-5.4 (Non-reasoning): 18.2 · $5.63/1MMuse Glimmer (high): 18.1 · $0.64/1MGemini 3 Flash Preview (Non-reasoning): 17.9 · $1.13/1MGrok 4 Fast (Reasoning): 17.9 · $0.28/1MQwen3.5 122B A10B (Non-reasoning): 17.7 · $1.10/1MClaude 4.5 Haiku (Reasoning): 17.6 · $2.00/1MGLM-4.7 (Non-reasoning): 17.4 · $1.00/1MGPT-5 mini (high): 17.4 · $0.69/1MRing-2.6-1T: 17.3 · $0.85/1MGPT-5.2 (Non-reasoning): 17.0 · $4.81/1MHy3-preview (Non-reasoning): 17.0 · $0.10/1MLing-2.6-1T: 17.0 · $0.85/1MStep 3.5 Flash 2603: 17.0 · $0.15/1MGemini 2.5 Pro: 16.7 · $3.44/1MGemma 4 26B A4B (Reasoning): 16.7 · $0.18/1Mo4-mini (high): 16.7 · $1.93/1MClaude 4 Opus (Non-reasoning): 16.6 · $30.00/1MDeepSeek V3.2 Exp (Reasoning): 16.6 · $0.32/1MStep 3.5 Flash: 16.6 · $0.15/1MQwen3 Max Thinking (Preview): 16.3 · $2.40/1MQwen3.5 122B A10B (Reasoning): 16.2 · $1.10/1MGPT-5.6 Luna (Non-reasoning): 16.1 · $0.45/1MDeepSeek V3.2 (Non-reasoning): 16.0 · $0.32/1MGemini 3.1 Flash-Lite: 16.0 · $0.56/1MQwen3 Max: 15.6 · $2.40/1MClaude 4.5 Haiku (Non-reasoning): 15.4 · $2.00/1MDeepSeek V3.1 Terminus (Reasoning): 15.4 · $1.91/1MClaude 3.7 Sonnet (Non-reasoning): 15.3 · $6.00/1MKimi K2 0905: 15.3 · $1.08/1Mo1: 15.2 · $26.25/1MQwen3.6 35B A3B (Non-reasoning): 15.2 · $0.84/1MQwen3.5 35B A3B (Non-reasoning): 15.1 · $0.69/1MGLM-4.6 (Non-reasoning): 14.9 · $0.98/1MGLM-4.7-Flash (Reasoning): 14.9 · $0.15/1MMistral Medium 3.5: 14.9 · $3.00/1MGranite 4.2 30B: 14.8 · $0.28/1MGrok 3 mini Reasoning (high): 14.6 · $0.35/1MGrok 4.20 0309 (Non-reasoning): 14.6 · $3.00/1MGemini 2.5 Pro Preview (May' 25): 14.5 · $3.44/1MGrok 4.3 (Non-reasoning): 14.5 · $1.56/1MGemma 4 12B (Reasoning): 14.2 · $0.15/1MGrok 4.20 0309 v2 (Non-reasoning): 14.2 · $1.56/1MNova 2.0 Pro Preview (medium): 14.2 · $3.44/1MDeepSeek V3.1 Terminus (Non-reasoning): 13.9 · $0.45/1MDeepSeek V3.2 Exp (Non-reasoning): 13.9 · $0.32/1MGemma 4 31B (Non-reasoning): 13.9 · $0.21/1MDeepSeek V3.1 (Non-reasoning): 13.7 · $0.85/1MQwen3.5 9B (Reasoning): 13.7 · $0.15/1MNemotron 3 Super 120B A12B (Reasoning): 13.6 · $0.31/1MNemotron 3.5 Lightning: 13.6 · $0.10/1MNova 2.0 Omni (medium): 13.6 · $0.85/1MDeepSeek V3.1 (Reasoning): 13.5 · $0.87/1MNova 2.0 Lite (high): 13.4 · $0.85/1MQwen3 VL 235B A22B (Reasoning): 13.4 · $1.30/1MGPT-5.1 (Non-reasoning): 13.3 · $3.44/1MQwen3.5 9B (Non-reasoning): 13.3 · $0.19/1MDeepSeek R1 0528 (May '25): 13.1 · $1.76/1MGemini 2.5 Flash (Reasoning): 13.1 · $0.85/1MGemma 4 26B A4B (Non-reasoning): 13.1 · $0.20/1MQwen3.5 4B (Reasoning): 13.1 · $0.06/1MGPT-5 nano (high): 13.0 · $0.14/1MNova 2.0 Pro Preview (low): 12.8 · $3.44/1MGPT-4.1: 12.7 · $3.50/1MKimi K2: 12.7 · $1.00/1MQwen3 235B A22B 2507 (Reasoning): 12.7 · $0.75/1MQwen3 Max (Preview): 12.6 · $2.40/1MGPT-5 nano (medium): 12.5 · $0.14/1MNova 2.0 Lite (medium): 12.5 · $0.85/1Mo3-mini: 12.5 · $1.93/1MQwen3.5 Omni Flash: 12.5 · $0.28/1Mo1-pro: 12.4 · $262.50/1Mgpt-oss-120b (high): 12.3 · $0.26/1MGrok 3: 12.1 · $8.00/1MSeed-OSS-36B-Instruct: 12.1 · $0.30/1MQwen3 235B A22B 2507 Instruct: 12.0 · $0.40/1MQwen3 Coder 480B A35B Instruct: 11.9 · $3.00/1MQwen3 VL 32B (Reasoning): 11.9 · $0.28/1MGranite 4.2 8B: 11.8 · $0.11/1MNova 2.0 Lite (low): 11.8 · $0.85/1MGPT-5.4 nano (Non-Reasoning): 11.7 · $0.46/1MMiniMax M1 80k: 11.7 · $0.96/1MMercury 2: 11.5 · $0.38/1MMistral Small 4 (Reasoning): 11.5 · $0.26/1MDeepSeek R1 (Jan '25): 11.4 · $2.50/1MGPT-5 (minimal): 11.4 · $3.44/1Mo1-preview: 11.4 · $28.88/1MGLM-4.6V (Reasoning): 11.2 · $0.45/1MQwen3 Next 80B A3B (Reasoning): 11.2 · $0.41/1MGLM-4.5-Air: 11.1 · $0.37/1MGPT-5.4 mini (Non-Reasoning): 11.1 · $1.69/1MGrok 4 Fast (Non-reasoning): 11.1 · $0.28/1MNova 2.0 Omni (low): 11.1 · $0.85/1Mo3-mini (high): 11.0 · $1.93/1MTrinity Large Thinking: 10.9 · $0.41/1MQwen3.5 4B (Non-reasoning): 10.8 · $0.06/1MGLM-4.7-Flash (Non-reasoning): 10.6 · $0.15/1MGemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning): 10.4 · $0.18/1MNemotron 3 Nano Omni 30B A3B Reasoning: 10.3 · $0.16/1MGPT-4.1 mini: 10.2 · $0.70/1Mgpt-oss-120b (low): 10.2 · $0.25/1MQwen3 Coder Next: 10.1 · $0.56/1Mgpt-oss-20b (low): 10.0 · $0.10/1MNova 2.0 Pro Preview (Non-reasoning): 10.0 · $3.44/1MGemini 2.5 Flash (Non-reasoning): 9.9 · $0.85/1MGPT-5 mini (minimal): 9.9 · $0.69/1MQwen3 VL 235B A22B Instruct: 9.9 · $0.70/1MQwen3 30B A3B 2507 (Reasoning): 9.8 · $0.75/1MDeepSeek V3 0324: 9.7 · $0.93/1MMistral Large 3: 9.7 · $0.75/1MQwen3 Coder 30B A3B Instruct: 9.6 · $0.90/1MQwen3 Next 80B A3B Instruct: 9.6 · $0.41/1MMistral Medium 3.1: 9.5 · $0.80/1MQwen3 235B A22B (Reasoning): 9.5 · $2.63/1MQwen3 VL 30B A3B (Reasoning): 9.5 · $0.75/1MQwQ 32B: 9.5 · $0.75/1MGemma 4 12B (Non-reasoning): 9.4 · $0.15/1MGemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning): 9.3 · $0.18/1MLlama 4 Maverick: 9.3 · $0.42/1MNova Premier: 9.2 · $5.00/1MGranite 4.2 3B: 9.1 · $0.05/1Mgpt-oss-20b (high): 9.0 · $0.09/1MLlama Nemotron Super 49B v1.5 (Reasoning): 9.0 · $0.40/1MMistral Medium 3: 9.0 · $0.80/1MMistral Small 4 (Non-reasoning): 9.0 · $0.26/1MGemma 4 E4B (Reasoning): 8.9 · $0.04/1MNVIDIA Nemotron 3 Nano 30B A3B (Reasoning): 8.9 · $0.09/1MSarvam 105B (high): 8.8 · $0.07/1MClaude 3 Opus: 8.7 · $30.00/1MNova 2.0 Lite (Non-reasoning): 8.7 · $0.85/1MMagistral Small 1.2: 8.6 · $0.75/1MDeepSeek V3 (Dec '24): 8.5 · $0.46/1MGemini 2.5 Flash-Lite (Reasoning): 8.5 · $0.18/1MGLM-4.6V (Non-reasoning): 8.4 · $0.45/1MGPT-4o (Nov '24): 8.4 · $4.38/1MQwen3 VL 32B Instruct: 8.4 · $0.28/1MQwen3 235B A22B (Non-reasoning): 8.3 · $1.23/1MNova 2.0 Omni (Non-reasoning): 8.2 · $0.85/1MQwen3 VL 8B (Reasoning): 8.2 · $0.66/1MClaude 3.5 Sonnet (Oct '24): 7.9 · $6.00/1MDeepSeek R1 Distill Llama 70B: 7.9 · $0.80/1MQwen3 VL 30B A3B Instruct: 7.9 · $0.35/1MGPT-4.1 nano: 7.8 · $0.18/1MLing-flash-2.0: 7.8 · $0.25/1MQwen3 Omni 30B A3B (Reasoning): 7.8 · $0.43/1MSolar Pro 3: 7.8 · $0.26/1MGPT-4o (Aug '24): 7.7 · $4.38/1MLlama 3.3 Instruct 70B: 7.7 · $0.71/1MQwen2.5 Instruct 72B: 7.7 · $0.48/1MGLM-4.5V (Reasoning): 7.6 · $0.90/1MQwen3 30B A3B (Reasoning): 7.6 · $0.75/1MERNIE 4.5 300B A47B: 7.5 · $0.49/1MGemma 4 E4B (Non-reasoning): 7.5 · $0.04/1MHermes 4 - Llama-3.1 405B (Reasoning): 7.5 · $1.50/1MNVIDIA Nemotron Nano 12B v2 VL (Reasoning): 7.5 · $0.30/1MQwen3 30B A3B 2507 Instruct: 7.5 · $0.35/1MHermes 4 - Llama-3.1 405B (Non-reasoning): 7.4 · $1.50/1MLlama Nemotron Super 49B v1.5 (Non-reasoning): 7.4 · $0.40/1MMistral Small 3.1: 7.4 · $0.15/1MNVIDIA Nemotron Nano 9B V2 (Reasoning): 7.4 · $0.07/1MGPT-4o (May '24): 7.3 · $7.50/1MQwen3 32B (Non-reasoning): 7.3 · $0.28/1MQwen3 VL 8B Instruct: 7.3 · $0.31/1MClaude 3.5 Sonnet (June '24): 7.2 · $6.00/1MQwen3 32B (Reasoning): 7.2 · $0.28/1MRing-flash-2.0: 7.2 · $0.25/1MGPT-5 nano (minimal): 7.1 · $0.14/1MCommand A: 7.0 · $4.38/1MGPT-4 Turbo: 7.0 · $15.00/1MMistral Small 3.2: 7.0 · $0.15/1MNova Pro: 7.0 · $1.40/1MLlama 3.1 Instruct 8B: 6.9 · $0.03/1MLlama 3.1 Nemotron Instruct 70B: 6.9 · $1.20/1MMistral Large 2 (Jul '24): 6.8 · $3.00/1MNVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning): 6.8 · $0.09/1MNVIDIA Nemotron Nano 9B V2 (Non-reasoning): 6.8 · $0.09/1MGemini 2.5 Flash-Lite (Non-reasoning): 6.7 · $0.18/1MGLM-4.5V (Non-reasoning): 6.7 · $0.90/1MGPT-4: 6.7 · $37.50/1MGPT-4o mini: 6.7 · $0.26/1MMistral Small 3: 6.7 · $0.15/1MNova Lite: 6.7 · $0.11/1MQwen3 14B (Non-reasoning): 6.7 · $0.61/1MGranite 4.1 8B: 6.6 · $0.06/1MLlama 3.1 Instruct 70B: 6.6 · $0.56/1MQwen3 30B A3B (Non-reasoning): 6.6 · $0.35/1MSarvam 30B (high): 6.6 · $0.05/1MLlama 4 Scout: 6.5 · $0.31/1MQwen2.5 Turbo: 6.4 · $0.09/1MQwen3 14B (Reasoning): 6.4 · $1.31/1MReka Flash (Sep '24): 6.4 · $0.35/1MSolar Mini: 6.4 · $0.15/1MCeleris-1: 6.3 · $0.33/1MGranite 4.0 H Small: 6.0 · $0.11/1MHermes 3 - Llama-3.1 70B: 6.0 · $0.70/1MJamba 1.5 Large: 6.0 · $3.50/1MMinistral 3 14B: 6.0 · $0.20/1MQwen3 8B (Non-reasoning): 6.0 · $0.31/1MQwen3 Omni 30B A3B Instruct: 6.0 · $0.43/1MNova Micro: 5.9 · $0.06/1MPhi-4: 5.9 · $0.22/1MMistral Large (Feb '24): 5.8 · $6.00/1MMistral Small (Sep '24): 5.8 · $0.30/1MNVIDIA Nemotron Nano 12B v2 VL (Non-reasoning): 5.8 · $0.30/1MLlama 2 Chat 7B: 5.7 · $0.10/1MClaude 3 Haiku: 5.6 · $0.50/1MReka Flash 3: 5.6 · $0.35/1MGPT-3.5 Turbo: 5.5 · $0.75/1MLlama 3 Instruct 70B: 5.5 · $1.18/1MMinistral 3 8B: 5.5 · $0.15/1MMistral Medium: 5.5 · $3.00/1MMistral Small (Feb '24): 5.5 · $0.26/1MLlama 3.2 Instruct 11B (Vision): 5.4 · $0.35/1MJamba 1.5 Mini: 5.2 · $0.25/1MOlmo 3 7B Instruct: 5.2 · $0.13/1MQwen3 8B (Reasoning): 5.2 · $0.66/1MApertus 70B Instruct: 5.1 · $1.35/1MMixtral 8x7B Instruct: 5.1 · $0.51/1MMistral 7B Instruct: 5.0 · $0.25/1MGranite 3.3 8B (Non-reasoning): 4.9 · $0.09/1MApertus 8B Instruct: 4.8 · $0.13/1MLlama 3 Instruct 8B: 4.8 · $0.07/1MMinistral 3 3B: 4.8 · $0.10/1M Claude Fable 5.1 (Adapt…Claude Fable 5.1 (Adapt…GPT-6 Astra (max)GPT-6 Astra (xhigh)Claude Fable 5.1 (Adapt…GPT-6 Astra (high)Claude Opus 5 (Adaptive…Claude Fable 5 (Adaptiv…Claude Opus 5 (Adaptive…GPT-6 Astra (medium)
  • Anthropic
  • OpenAI
  • Meta
  • Alibaba
  • Z.ai
  • xAI
  • Kimi

Output speed

Median output tokens per second · higher is better

Artificial Analysis
Muse Spark 1.3 (max): 234234Muse Spark 1.3 (max)Muse Spark 1.3 (xhigh): 223223Muse Spark 1.3 (xhigh)Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback): 7373Claude Fable 5.1 (Adaptiv…GLM-5.3 (max): 7272GLM-5.3 (max)GPT-5.6 Sol (max): 6565GPT-5.6 Sol (max)Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback): 6464Claude Fable 5 (Adaptive …Grok 4.6 (xhigh): 6161Grok 4.6 (xhigh)GPT-5.6 Sol (xhigh): 6060GPT-5.6 Sol (xhigh)GPT-6 Astra (max): 5959GPT-6 Astra (max)GPT-6 Astra (xhigh): 5858GPT-6 Astra (xhigh)Grok 4.6 (high): 5757Grok 4.6 (high)Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback): 5757Claude Fable 5.1 (Adaptiv…Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback): 5555Claude Fable 5.1 (Adaptiv…GPT-6 Astra (low): 5454GPT-6 Astra (low)GPT-6 Astra (high): 5454GPT-6 Astra (high)Claude Opus 5 (Adaptive Reasoning, Max Effort): 5454Claude Opus 5 (Adaptive R…GPT-6 Astra (medium): 5353GPT-6 Astra (medium)Claude Opus 5 (Adaptive Reasoning, Xhigh Effort): 5353Claude Opus 5 (Adaptive R…Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback): 5151Claude Fable 5.1 (Adaptiv…Claude Opus 5 (Adaptive Reasoning, High Effort): 5050Claude Opus 5 (Adaptive R…

Latency

Median seconds to first token · lower is better

Artificial Analysis
GPT-6 Astra (low): 1.33s1.33sGPT-6 Astra (low)Qwen3.8 Max (0902): 1.66s1.66sQwen3.8 Max (0902)GLM-5.3 (max): 2.65s2.65sGLM-5.3 (max)Kimi K3 (max): 3.18s3.18sKimi K3 (max)Claude Opus 5 (Adaptive Reasoning, Medium Effort): 3.40s3.40sClaude Opus 5 (Adaptive R…Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback): 4.02s4.02sClaude Fable 5.1 (Adaptiv…GPT-6 Astra (medium): 4.05s4.05sGPT-6 Astra (medium)Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback): 5.91s5.91sClaude Fable 5.1 (Adaptiv…Claude Opus 5 (Adaptive Reasoning, High Effort): 7.84s7.84sClaude Opus 5 (Adaptive R…Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback): 8.64s8.64sClaude Fable 5.1 (Adaptiv…Claude Opus 5 (Adaptive Reasoning, Xhigh Effort): 17.59s17.59sClaude Opus 5 (Adaptive R…Muse Spark 1.3 (xhigh): 20.72s20.72sMuse Spark 1.3 (xhigh)Muse Spark 1.3 (max): 23.42s23.42sMuse Spark 1.3 (max)Claude Opus 5 (Adaptive Reasoning, Max Effort): 25.73s25.73sClaude Opus 5 (Adaptive R…GPT-6 Astra (high): 26.87s26.87sGPT-6 Astra (high)GPT-5.6 Sol (xhigh): 27.34s27.34sGPT-5.6 Sol (xhigh)Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback): 30.35s30.35sClaude Fable 5.1 (Adaptiv…Grok 4.6 (high): 31.81s31.81sGrok 4.6 (high)Grok 4.6 (xhigh): 35.96s35.96sGrok 4.6 (xhigh)Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback): 44.13s44.13sClaude Fable 5 (Adaptive …

Price

USD per 1M tokens for each lab's newest flagship-class model · input (left) and output (right) · lower is better

OpenRouter · live
Phi 4: $0.07$0.07Phi 4: $0.14$0.14Phi 4Nemotron 3.5 Lightning: $0.07$0.07Nemotron 3.5 Lightning: $0.2$0.2Nemotron 3.5 LightningDeepSeek V4 Pro 0813: $0.58$0.58DeepSeek V4 Pro 0813: $1.73$1.73DeepSeek V4 Pro 0813GLM 5.3: $0.91$0.91GLM 5.3: $2.86$2.86GLM 5.3Muse Spark 1.3: $1.25$1.25Muse Spark 1.3: $4.25$4.25Muse Spark 1.3Qwen3.8 Max (0902): $2.00$2.00Qwen3.8 Max (0902): $6.00$6.00Qwen3.8 Max (0902)Grok 4.6: $2.00$2.00Grok 4.6: $6.00$6.00Grok 4.6Mistral Medium 3.5: $1.50$1.50Mistral Medium 3.5: $7.50$7.50Mistral Medium 3.5Kimi K3: $1.70$1.70Kimi K3: $8.50$8.50Kimi K3Command A: $2.50$2.50Command A: $10$10Command ANova Premier 1.0: $2.50$2.50Nova Premier 1.0: $13$13Nova Premier 1.0GPT-6 Astra: $10$10GPT-6 Astra: $50$50GPT-6 AstraClaude Fable 5.1: $10$10Claude Fable 5.1: $50$50Claude Fable 5.1
  • Microsoft
  • NVIDIA
  • DeepSeek
  • Z.ai
  • Meta
  • Alibaba
  • xAI
  • Mistral
  • Moonshot
  • Cohere
  • Amazon
  • OpenAI
  • Anthropic

Context window

Tokens of input a model can read at once · larger is better

OpenRouter · live
Grok 4.20: 2M2MGrok 4.20GLM 5.3 Flash: 1.31M1.31MGLM 5.3 FlashDeepSeek V4 Flash 0731: 1.31M1.31MDeepSeek V4 Flash 0731Llama 4 Scout: 1.31M1.31MLlama 4 ScoutGPT-6 Astra: 1.05M1.05MGPT-6 AstraGLM 5.3 FlashX: 1.05M1.05MGLM 5.3 FlashXDeepSeek V4.1 Flash: 1.05M1.05MDeepSeek V4.1 FlashGemini 3.8 Flash: 1.05M1.05MGemini 3.8 FlashMuse Spark 1.3: 1.05M1.05MMuse Spark 1.3Qwen3.8 2.4T A95B: 1.05M1.05MQwen3.8 2.4T A95BKimi K3: 1.05M1.05MKimi K3MiniMax M3: 1.05M1.05MMiniMax M3GPT-4.1: 1.05M1.05MGPT-4.1MiniMax-01: 1M1MMiniMax-01Qwen3.8 Max (0902): 1M1MQwen3.8 Max (0902)Claude Fable 5.1: 1M1MClaude Fable 5.1
  • xAI
  • Z.ai
  • DeepSeek
  • Meta
  • OpenAI
  • Google
  • Alibaba
  • Moonshot
  • MiniMax
  • Anthropic

Releases over time

One dot per model release in the last twelve months · most active labs

OpenRouter · live
OctNovDecJanFebMarAprMayJunJulAugSep OpenAI31 releases GPT-6 Astra · 4 Sep 2026GPT-6 Astra Pro · 4 Sep 2026GPT-5.6 Luna · 9 Jul 2026GPT-5.6 Luna Pro · 9 Jul 2026GPT-5.6 Sol · 9 Jul 2026GPT-5.6 Sol Pro · 9 Jul 2026GPT-5.6 Terra · 9 Jul 2026GPT-5.6 Terra Pro · 9 Jul 2026GPT Chat Latest · 5 May 2026GPT-5.5 · 24 Apr 2026GPT-5.5 Pro · 24 Apr 2026GPT-5.4 Image 2 · 21 Apr 2026GPT-5.4 Mini · 17 Mar 2026GPT-5.4 Nano · 17 Mar 2026GPT-5.4 · 5 Mar 2026GPT-5.4 Pro · 5 Mar 2026GPT-5.3-Codex · 24 Feb 2026GPT Audio · 19 Jan 2026GPT Audio Mini · 19 Jan 2026GPT-5.2-Codex · 14 Jan 2026GPT-5.2 · 10 Dec 2025GPT-5.2 Chat · 10 Dec 2025GPT-5.2 Pro · 10 Dec 2025GPT-5.1-Codex-Max · 4 Dec 2025GPT-5.1 · 13 Nov 2025GPT-5.1-Codex · 13 Nov 2025GPT-5.1-Codex-Mini · 13 Nov 2025gpt-oss-safeguard-20b · 29 Oct 2025GPT-5 Image Mini · 16 Oct 2025GPT-5 Image · 14 Oct 2025GPT-5 Pro · 6 Oct 2025 Alibaba31 releases Qwen3.8 Max (0902) · 3 Sep 2026Qwen3.8 Flash · 26 Aug 2026Qwen3.8 27B · 14 Aug 2026Qwen3.8 2.4T A95B · 12 Aug 2026Qwen3.7 Flash · 27 Jul 2026Qwen3.7 Plus · 3 Jun 2026Qwen3.7 Max · 21 May 2026Qwen3.5 Plus 2026-04-20 · 27 Apr 2026Qwen3.6 27B · 27 Apr 2026Qwen3.6 35B A3B · 27 Apr 2026Qwen3.6 Flash · 27 Apr 2026Qwen3.6 Max Preview · 27 Apr 2026Qwen3.6 Plus · 2 Apr 2026Qwen3.5-9B · 10 Mar 2026Qwen3.5-122B-A10B · 25 Feb 2026Qwen3.5-27B · 25 Feb 2026Qwen3.5-35B-A3B · 25 Feb 2026Qwen3.5-Flash · 25 Feb 2026Qwen3.5 397B A17B · 16 Feb 2026Qwen3.5 Plus 2026-02-15 · 16 Feb 2026Qwen3 Max Thinking · 9 Feb 2026Qwen3 Coder Next · 4 Feb 2026Qwen3 VL 32B Instruct · 23 Oct 2025Qwen3 VL 8B Instruct · 14 Oct 2025Qwen3 VL 8B Thinking · 14 Oct 2025Qwen3 VL 30B A3B Instruct · 6 Oct 2025Qwen3 VL 30B A3B Thinking · 6 Oct 2025Qwen3 Coder Plus · 23 Sep 2025Qwen3 Max · 23 Sep 2025Qwen3 VL 235B A22B Instruct · 23 Sep 2025Qwen3 VL 235B A22B Thinking · 23 Sep 2025 Google20 releases Gemini 3.8 Flash · 2 Sep 2026Gemini 3.7 Flash · 13 Aug 2026Gemini 3.5 Flash Lite · 21 Jul 2026Gemini 3.6 Flash · 21 Jul 2026Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) · 30 Jun 2026Nano Banana 2 (Gemini 3.1 Flash Image) · 18 Jun 2026Nano Banana Pro (Gemini 3 Pro Image) · 18 Jun 2026Gemini 3.5 Flash · 19 May 2026Gemini 3.1 Flash Lite · 7 May 2026Gemma 4 26B A4B · 3 Apr 2026Gemma 4 31B · 2 Apr 2026Lyria 3 Clip Preview · 30 Mar 2026Lyria 3 Pro Preview · 30 Mar 2026Gemini 3.1 Flash Lite Preview · 3 Mar 2026Nano Banana 2 (Gemini 3.1 Flash Image Preview) · 26 Feb 2026Gemini 3.1 Pro Preview Custom Tools · 25 Feb 2026Gemini 3.1 Pro Preview · 19 Feb 2026Gemini 3 Flash Preview · 17 Dec 2025Nano Banana Pro (Gemini 3 Pro Image Preview) · 20 Nov 2025Nano Banana (Gemini 2.5 Flash Image) · 7 Oct 2025 Z.ai12 releases GLM 5.3 FlashX · 18 Sep 2026GLM 5.3 Flash · 26 Aug 2026GLM 5.3 · 18 Aug 2026GLM 5.2 · 16 Jun 2026GLM 5.1 · 7 Apr 2026GLM 5V Turbo · 1 Apr 2026GLM 5 Turbo · 15 Mar 2026GLM 5 · 11 Feb 2026GLM 4.7 Flash · 19 Jan 2026GLM 4.7 · 22 Dec 2025GLM 4.6V · 8 Dec 2025GLM 4.6 · 30 Sep 2025 Anthropic11 releases Claude Fable 5.1 · 1 Sep 2026Claude Opus 5 · 24 Jul 2026Claude Sonnet 5 · 30 Jun 2026Claude Fable 5 · 9 Jun 2026Claude Opus 4.8 · 27 May 2026Claude Opus 4.7 · 16 Apr 2026Claude Sonnet 4.6 · 17 Feb 2026Claude Opus 4.6 · 4 Feb 2026Claude Opus 4.5 · 24 Nov 2025Claude Haiku 4.5 · 15 Oct 2025Claude Sonnet 4.5 · 29 Sep 2025 DeepSeek9 releases DeepSeek V4.1 Flash · 10 Sep 2026DeepSeek V4 Flash Vision Exp · 21 Aug 2026DeepSeek V4 Pro 0813 · 12 Aug 2026DeepSeek V4 Flash 0731 · 31 Jul 2026DeepSeek V4 Flash 0423 · 24 Apr 2026DeepSeek V4 Pro 0423 · 24 Apr 2026DeepSeek V3.2 · 1 Dec 2025DeepSeek V3.2 Exp · 29 Sep 2025DeepSeek V3.1 Terminus · 22 Sep 2025 Mistral7 releases Mistral Medium 3.5 · 30 Apr 2026Mistral Small 4 · 16 Mar 2026Devstral 2 2512 · 9 Dec 2025Ministral 3 14B 2512 · 2 Dec 2025Ministral 3 3B 2512 · 2 Dec 2025Ministral 3 8B 2512 · 2 Dec 2025Voxtral Small 24B 2507 · 30 Oct 2025 Meta6 releases Muse Spark 1.3 · 2 Sep 2026Muse Spark 1.3 Contributor · 2 Sep 2026Muse Spark 1.2 Contributor · 21 Aug 2026Muse Glimmer 30B · 9 Aug 2026Muse Spark 1.2 · 5 Aug 2026Muse Spark 1.1 · 16 Jul 2026 xAI6 releases Grok 4.6 · 12 Aug 2026Grok 4.5 · 8 Jul 2026Grok Build 0.1 · 20 May 2026Grok 4.3 · 30 Apr 2026Grok 4.20 · 31 Mar 2026Grok 4.20 Multi-Agent · 31 Mar 2026 MiniMax6 releases MiniMax M3 · 31 May 2026MiniMax M2.7 · 18 Mar 2026MiniMax M2.5 · 12 Feb 2026MiniMax M2-her · 23 Jan 2026MiniMax M2.1 · 23 Dec 2025MiniMax M2 · 23 Oct 2025

Open weights vs. proprietary

Releases in the last twelve months per lab · open weights can be self-hosted

OpenRouter · live
  1. OpenAI1 open · 30 closed
  2. Alibaba18 open · 13 closed
  3. Google2 open · 18 closed
  4. Z.ai9 open · 3 closed
  5. Anthropic0 open · 11 closed
  6. DeepSeek9 open · 0 closed
  7. Mistral6 open · 1 closed
  8. Meta1 open · 5 closed
  9. xAI0 open · 6 closed
  10. MiniMax5 open · 1 closed
47%of all 327 tracked models ship open weightsOpen weights Proprietary

Compare the frontier

Each lab's newest flagship-class model

OpenRouter · live
ModelLabReleasedContextInput $/1MOutput $/1MCapabilities
GPT-6 AstraOpenAI4 Sep 20261.05M$10$50
ReasoningVision
Claude Fable 5.1Anthropic1 Sep 20261M$10$50
ReasoningVision
DeepSeek V4 Pro 0813DeepSeek12 Aug 20261.05M$0.58$1.73
ReasoningOpen weights
Qwen3.8 Max (0902)Alibaba3 Sep 20261M$2.00$6.00
ReasoningVisionVideo
Grok 4.6xAI12 Aug 2026500K$2.00$6.00
ReasoningVision
Muse Spark 1.3Meta2 Sep 20261.05M$1.25$4.25
ReasoningVisionAudioVideo
Mistral Medium 3.5Mistral30 Apr 2026262K$1.50$7.50
ReasoningVision
GLM 5.3Z.ai18 Aug 20261.31M$0.91$2.86
ReasoningOpen weights
Kimi K3Moonshot16 Jul 20261.05M$1.70$8.50
ReasoningVisionVideoOpen weights
Nova Premier 1.0Amazon31 Oct 20251M$2.50$13
Vision
Nemotron 3.5 LightningNVIDIA11 Aug 2026262K$0.07$0.2
ReasoningOpen weights
Phi 4Microsoft10 Jan 202516K$0.07$0.14
Open weights
Command ACohere13 Mar 2025256K$2.50$10
Open weights

Benchmark data: Artificial Analysis (synced 13 min ago). Catalogue, pricing and context data: OpenRouter (synced 13 min ago). Prices are list prices per million tokens in USD and change without notice.