Skip to content

Mistral · Model profile

Mistral Small 4

Intelligence, Performance & Price Analysis

Source: Canonical slug: mistral-small-4 · Canonical model registry at build time

Routing verdict

Strong general-purpose model

Suitable for most production tasks, but high-volume or repetitive work should still be compared against cheaper routes.

Mistral Small 4 summary metrics

Intelligence
20 scoreMistral Small 4 (Reasoning) scores 20 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).
Speed
177.4 output tokens/secMistral Small 4 (Reasoning) generates output at 177.4 tokens per second (based on Mistral's API), which is well above average compared to other open weight models of similar size (median: 88.2 t/s).
Latency
0.78s TTFTMistral Small 4 (Reasoning) has a time to first token (TTFT) of 0.78s (based on Mistral's API), which is very competitive compared to other open weight models of similar size (median: 1.52s).
Input Price
$0.15 / 1M tokensMistral Small 4 (Reasoning) costs $0.15 per 1M input tokens (very competitive, median: $0.40) and $0.60 per 1M output tokens (better than average, median: $0.84), based on Mistral's API.
Output Price
$0.60 / 1M tokensMistral Small 4 (Reasoning) costs $0.15 per 1M input tokens (very competitive, median: $0.40) and $0.60 per 1M output tokens (better than average, median: $0.84), based on Mistral's API.
Verbosity
53M Output tokens from Intelligence Index 2 out of 4 units for Verbosity . Compa

Assessment

Routing verdict and task fit

Mistral Small 4 (Reasoning) scores 20 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9). Mistral Small 4 (Reasoning) generates output at 177.4 tokens per second (based on Mistral's API), which is well above average compared to other open weight models of similar size (median: 88.2 t/s). Mistral Small 4 (Reasoning) costs $0.15 per 1M input tokens (very competitive, median: $0.40) and $0.60 per 1M output tokens (better than average, median: $0.84), based on Mistral's API.
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsOptimalIntelligence score 20 supports capable reasoning, but very hard tasks may benefit from higher-tier models.
High-volume chat & customer-facingOptimalOutput speed 177.4 tokens/sec and capable intelligence make this suitable for real-time chat at scale.
Latency-sensitive applicationsOptimalTTFT 0.78s — among the lowest latencies, suitable for interactive latency-critical use cases.
Cost-sensitive pipelinesOptimalOutput pricing at $0.60 is reasonable for moderate volume.

Evidence

Evidence charts

17 published comparison charts for this profile. Mistral Small 4 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. Mistral Small 4: -29.9, row 11 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 K2.66.4
  6. Muse Spark4.1
  7. GLM-5.2 (max)4
  8. MiniMax-M31.4
  9. Nemotron 3 Ultra-0.833
  10. DeepSeek V4 Pro (max)-10
  11. Mistral Small 4 (this model)-29.9
  12. K2 Think V2-33.9
  13. Mistral Medium 3.5-36.3
  14. Qwen3.5 122B A10B-39.6
  15. NVIDIA Nemotron 3 Super-42.1
  16. Devstral 2-46.2
  17. gpt-oss-120b (high)-50
  18. Qwen3 Next 80B A3B-50.6
  19. Llama 3.3 70B-51.9
  20. Llama 4 Scout-52.4

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. Mistral Small 4: 19.6, row 16 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 K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. Qwen3.5 122B A10B32.3
  12. Mistral Medium 3.529.9
  13. NVIDIA Nemotron 3 Super25.4
  14. gpt-oss-120b (high)23.8
  15. Qwen3 Coder Next21.1
  16. Mistral Small 4 (this model)19.6
  17. Devstral 219.2
  18. HyperNova 60B 260517.8
  19. K2 Think V217.3
  20. Qwen3 Next 80B A3B16.7

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. Mistral Small 4: 19.6, row 16 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 K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. Qwen3.5 122B A10B32.3
  12. Mistral Medium 3.529.9
  13. NVIDIA Nemotron 3 Super25.4
  14. gpt-oss-120b (high)23.8
  15. Qwen3 Coder Next21.1
  16. Mistral Small 4 (this model)19.6
  17. Devstral 219.2
  18. HyperNova 60B 260517.8
  19. K2 Think V217.3
  20. Qwen3 Next 80B A3B16.7

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

17 models in the published order. Mistral Small 4: 38.9, row 8 of 17 in the published order.
  1. K2 Think V288.9
  2. Nemotron 3 Ultra83.3
  3. NVIDIA Nemotron 3 Super83.3
  4. DeepSeek V4 Pro (max)50
  5. GLM-5.2 (max)44.4
  6. Qwen3 Next 80B A3B44.4
  7. Qwen3 Coder Next41.7
  8. Mistral Small 4 (this model)38.9
  9. gpt-oss-120b (high)38.9
  10. Qwen3.5 122B A10B38.9
  11. HyperNova 60B 260538.9
  12. Ling 2.6 Flash38.9
  13. Llama 3.3 70B38.9
  14. Kimi K2.633.3
  15. MiniMax-M333.3
  16. Mistral Medium 3.533.3
  17. Llama 4 Scout27.8

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

12 models in the published order. Mistral Small 4: 19.6, row 12 of 12 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 K2.644.2
  8. Muse Spark43.1
  9. Nemotron 3 Ultra37.8
  10. Grok 4.3 (high)37.6
  11. gpt-oss-120b (high)23.8
  12. Mistral Small 4 (this model)19.6

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.

Output Speed

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

20 models in the published order. Mistral Small 4: 177 t/s, row 7 of 20 in the published order.
  1. NVIDIA Nemotron 3 Super364 t/s
  2. HyperNova 60B 2605353 t/s
  3. gpt-oss-120b (high)314 t/s
  4. Nemotron 3 Ultra250 t/s
  5. GLM-5.2 (max)218 t/s
  6. Gemini 3.5 Flash192 t/s
  7. Mistral Small 4 (this model)177 t/s
  8. Ling 2.6 Flash177 t/s
  9. Grok 4.3 (high)164 t/s
  10. Mistral Medium 3.5140 t/s
  11. Qwen3.5 122B A10B139 t/s
  12. Qwen3 Next 80B A3B115 t/s
  13. Llama 4 Scout109 t/s
  14. MiniMax-M399 t/s
  15. Qwen3 Coder Next89 t/s
  16. GPT-5.5 (xhigh)88 t/s
  17. Llama 3.3 70B86 t/s
  18. Devstral 276 t/s
  19. Kimi K2.676 t/s
  20. DeepSeek V4 Pro (max)72 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

11 models in the published order. Mistral Small 4: 177 t/s, row 5 of 11 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. Mistral Small 4 (this model)177 t/s
  6. Grok 4.3 (high)164 t/s
  7. MiniMax-M399 t/s
  8. GPT-5.5 (xhigh)88 t/s
  9. Kimi K2.676 t/s
  10. DeepSeek V4 Pro (max)72 t/s
  11. 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. Mistral Small 4: 2.82 s, row 11 of 20 in the published order.
  1. Ling 2.6 Flash2.82 s
  2. Llama 4 Scout4.61 s
  3. Llama 3.3 70B5.85 s
  4. Qwen3 Coder Next5.65 s
  5. Devstral 26.55 s
  6. HyperNova 60B 26051.42 s
  7. NVIDIA Nemotron 3 Super1.37 s
  8. gpt-oss-120b (high)1.59 s
  9. Nemotron 3 Ultra2 s
  10. GLM-5.2 (max)2.3 s
  11. Mistral Small 4 (this model)2.82 s
  12. Mistral Medium 3.53.56 s
  13. Qwen3.5 122B A10B3.59 s
  14. Qwen3 Next 80B A3B4.35 s
  15. Grok 4.3 (high)3.05 s
  16. MiniMax-M35.05 s
  17. Gemini 3.5 Flash2.6 s
  18. Kimi K2.66.56 s
  19. DeepSeek V4 Pro (max)6.97 s
  20. GPT-5.5 (xhigh)5.68 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. Mistral Small 4: 11.27 s, row 11 of 20 in the published order.
  1. Llama 4 Scout0 s
  2. Devstral 20 s
  3. Ling 2.6 Flash0 s
  4. Llama 3.3 70B0 s
  5. Qwen3 Coder Next0 s
  6. HyperNova 60B 26055.67 s
  7. NVIDIA Nemotron 3 Super5.5 s
  8. gpt-oss-120b (high)6.37 s
  9. Nemotron 3 Ultra9.09 s
  10. GLM-5.2 (max)9.19 s
  11. Mistral Small 4 (this model)11.27 s
  12. Mistral Medium 3.514.25 s
  13. Qwen3.5 122B A10B14.35 s
  14. Qwen3 Next 80B A3B17.41 s
  15. Grok 4.3 (high)0 s
  16. MiniMax-M320.19 s
  17. Gemini 3.5 Flash0 s
  18. Kimi K2.658.43 s
  19. DeepSeek V4 Pro (max)60.99 s
  20. GPT-5.5 (xhigh)0 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. Mistral Small 4: 256k, row 16 of 20 in the published order.
  1. Llama 4 Scout10M
  2. Gemini 3.5 Flash1M
  3. Claude Fable 5 (with fallback)1M
  4. GLM-5.2 (max)1M
  5. DeepSeek V4 Pro (max)1M
  6. Grok 4.3 (high)1M
  7. MiniMax-M31M
  8. NVIDIA Nemotron 3 Super1M
  9. GPT-5.5 (xhigh)922k
  10. Nemotron 3 Ultra262.1k
  11. Muse Spark262.1k
  12. Qwen3.5 122B A10B262.1k
  13. K2 Think V2262.1k
  14. Qwen3 Next 80B A3B262.1k
  15. Ling 2.6 Flash262.1k
  16. Mistral Small 4 (this model)256k
  17. Kimi K2.6256k
  18. Mistral Medium 3.5256k
  19. Qwen3 Coder Next256k
  20. Devstral 2256k

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

18 models in the published order. Mistral Small 4 is not marked in this published set of 18 models.
  1. Llama 4 Scout0.000824
  2. HyperNova 60B 26050.000717
  3. DeepSeek V4 Pro (max)0.0063
  4. gpt-oss-120b (high)0.00454
  5. Llama 3.3 70B0.0012
  6. MiniMax-M30.0139
  7. Grok 4.3 (high)0.0101
  8. Qwen3 Next 80B A3B0.0111
  9. NVIDIA Nemotron 3 Super0.00577
  10. Qwen3.5 122B A10B0.0208
  11. Nemotron 3 Ultra0.0209
  12. Qwen3 Coder Next0.0184
  13. Kimi K2.60.0216
  14. GLM-5.2 (max)0.0268
  15. Gemini 3.5 Flash0.0884
  16. GPT-5.5 (xhigh)0.107
  17. Mistral Medium 3.50.0529
  18. 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. Mistral Small 4 is not marked in this published set of 10 models.
  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$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

18 models in the published order. Mistral Small 4 is not marked in this published set of 18 models.
  1. Llama 4 Scout2.7
  2. HyperNova 60B 26050.947
  3. Llama 3.3 70B2.7
  4. gpt-oss-120b (high)3.5
  5. DeepSeek V4 Pro (max)7.2
  6. MiniMax-M314.5
  7. Qwen3 Coder Next38.5
  8. NVIDIA Nemotron 3 Super5.4
  9. Grok 4.3 (high)12.5
  10. Qwen3 Next 80B A3B15.9
  11. Nemotron 3 Ultra33.5
  12. Qwen3.5 122B A10B32.7
  13. GLM-5.2 (max)32.9
  14. Kimi K2.625
  15. Gemini 3.5 Flash114
  16. Mistral Medium 3.553.7
  17. GPT-5.5 (xhigh)131.8
  18. 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. Mistral Small 4: $0.750, row 3 of 20 in the published order.
  • Input
  • Output
  • Cache Hit
  1. HyperNova 60B 2605$0.180
  2. Ling 2.6 Flash$0.420
  3. Mistral Small 4 (this model)$0.750
  4. gpt-oss-120b (high)$0.900
  5. Llama 4 Scout$0.830
  6. NVIDIA Nemotron 3 Super$1.17
  7. Llama 3.3 70B$1.87
  8. DeepSeek V4 Pro (max)$1.31
  9. MiniMax-M3$1.56
  10. Qwen3 Coder Next$1.90
  11. Nemotron 3 Ultra$3.60
  12. Qwen3.5 122B A10B$3.60
  13. Grok 4.3 (high)$3.95
  14. Kimi K2.6$5.11
  15. GLM-5.2 (max)$6.06
  16. Qwen3 Next 80B A3B$6.50
  17. Mistral Medium 3.5$9.15
  18. Gemini 3.5 Flash$10.65
  19. GPT-5.5 (xhigh)$35.50
  20. Claude Fable 5 (with fallback)$61.00

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. Mistral Small 4: 1.7, row 9 of 20 in the published order.
  1. Llama 4 Scout0.198
  2. Llama 3.3 70B0.321
  3. NVIDIA Nemotron 3 Super1.1
  4. Ling 2.6 Flash1.1
  5. HyperNova 60B 26051.2
  6. Devstral 21.3
  7. Nemotron 3 Ultra1.4
  8. Grok 4.3 (high)1.6
  9. Mistral Small 4 (this model)1.7
  10. Qwen3.5 122B A10B1.8
  11. gpt-oss-120b (high)1.9
  12. Gemini 3.5 Flash2.2
  13. Qwen3 Coder Next2.6
  14. Qwen3 Next 80B A3B2.8
  15. GLM-5.2 (max)2.8
  16. Mistral Medium 3.52.9
  17. GPT-5.5 (xhigh)3
  18. MiniMax-M33.9
  19. Claude Fable 5 (with fallback)4.8
  20. DeepSeek V4 Pro (max)6.8

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.

Model Size: Total and Active Parameters

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

18 models; segments: Passive Params, Active Params. Mistral Small 4: 119B, row 10 of 18 in the published order.
  • Passive Params
  • Active Params
  1. DeepSeek V4 Pro (max)1,600B
  2. Kimi K2.61,000B
  3. GLM-5.2 (max)753B
  4. Nemotron 3 Ultra550B
  5. MiniMax-M3428B
  6. Mistral Medium 3.5128B
  7. Qwen3.5 122B A10B125B
  8. Devstral 2125B
  9. NVIDIA Nemotron 3 Super120.6B
  10. Mistral Small 4 (this model)119B
  11. gpt-oss-120b (high)117B
  12. Llama 4 Scout109B
  13. Ling 2.6 Flash107B
  14. Qwen3 Next 80B A3B80B
  15. Qwen3 Coder Next79.7B
  16. K2 Think V270B
  17. Llama 3.3 70B70B
  18. HyperNova 60B 260558.7B

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. Mistral Small 4: 6,364.2, row 8 of 20 in the published order.
  1. Llama 4 Scout1,249.2
  2. Llama 3.3 70B1,685.9
  3. Devstral 25,946.4
  4. Muse Spark2,857.2
  5. Ling 2.6 Flash13,070.1
  6. Grok 4.3 (high)4,032
  7. Qwen3 Coder Next15,355.4
  8. Mistral Small 4 (this model)6,364.2
  9. GPT-5.5 (xhigh)3,572.5
  10. Qwen3.5 122B A10B6,491.8
  11. Qwen3 Next 80B A3B1,857.4
  12. Nemotron 3 Ultra7,796.7
  13. MiniMax-M311,623.4
  14. Mistral Medium 3.57,048.7
  15. NVIDIA Nemotron 3 Super7,442.1
  16. Gemini 3.5 Flash9,819.4
  17. HyperNova 60B 26055,120
  18. Claude Fable 5 (with fallback)7,695.8
  19. gpt-oss-120b (high)7,571.4
  20. DeepSeek V4 Pro (max)7,243.2

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

Mistral Small 4 technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningYesfaq
Input modalitiesMistral Small 4 (Reasoning) supports text and image input.faq
Output modalitiesMistral Small 4 (Reasoning) supports text output.faq
Context window260k tokensfaq
Open weights / sourceYes, Mistral Small 4 (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersMistral Small 4 (Reasoning) has 119 billion parameters (6.5 billion active).faq
Active parametersMistral Small 4 (Reasoning) is a Mixture of Experts (MoE) model with 119 billion total parameters, but only 6.5 billion active parameters are used during inference.faq
LicenseMistral Small 4 (Reasoning) is released under the Apache 2.0 license. This license allows commercial use.faq
API availabilityYes, Mistral Small 4 (Reasoning) is available via API through 1 provider.faq

Method

Methodology & Provenance

This page is rendered from the normalized profile and page JSON for Mistral Small 4.

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

Questions

Model FAQ and technical disclosures

Mistral Small 4 (Reasoning) was released on March 16, 2026.

Mistral Small 4 (Reasoning) was created by Mistral.

Mistral Small 4 (Reasoning) scores 20 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 9).

Mistral Small 4 (Reasoning) generates output at 177.4 tokens per second (based on Mistral's API), which is well above average compared to other open weight models of similar size (median: 88.2 t/s).

Mistral Small 4 (Reasoning) has a time to first token (TTFT) of 0.78s (based on Mistral's API), which is very competitive compared to other open weight models of similar size (median: 1.52s).

Mistral Small 4 (Reasoning) costs $0.15 per 1M input tokens (very competitive, median: $0.40) and $0.60 per 1M output tokens (better than average, median: $0.84), based on Mistral's API.

Mistral Small 4 (Reasoning) costs $0.15 per 1M input tokens and $0.60 per 1M output tokens (based on Mistral's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $0.20 per 1M tokens. Pricing may vary by provider.

When evaluated on the Intelligence Index, Mistral Small 4 (Reasoning) generated 53M output tokens, which is better than average compared to other open weight models of similar size (median: 59M).

Yes, Mistral Small 4 (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.

Mistral Small 4 (Reasoning) supports text and image input.

Mistral Small 4 (Reasoning) supports text output.

Yes, Mistral Small 4 (Reasoning) supports image input and can analyze, describe, and answer questions about images.

Yes, Mistral Small 4 (Reasoning) is multimodal. It can process text and image input and generate text output.

Mistral Small 4 (Reasoning) has a context window of 260k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Mistral Small 4 (Reasoning) is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Mistral Small 4 (Reasoning) has 119 billion parameters (6.5 billion active).

Mistral Small 4 (Reasoning) is a Mixture of Experts (MoE) model with 119 billion total parameters, but only 6.5 billion active parameters are used during inference.

Mistral Small 4 (Reasoning) is released under the Apache 2.0 license. This license allows commercial use.

Mistral Small 4 (Reasoning) achieves a score of 20 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Mistral Small 4 (Reasoning) is available via API through 1 provider.

Mistral Small 4 (Reasoning) is available through 1 API provider.

Models

  • 01MistralOther Mistral models in the catalog.
  • 02LeaderboardsCertified metric ranks, when published.