Skip to content

Mistral · Model profile

Mistral Small (Sep)

Intelligence, Performance & Price Analysis

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

Routing verdict

Budget-friendly / task-specific model

Best for high-volume, simple, or domain-specific tasks where cost or speed matters more than deep reasoning.

Mistral Small (Sep) summary metrics

Intelligence
5 scoreMistral Small (Sep '24) scores 5 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 6).
Speed
162.5 output tokens/secMistral Small (Sep '24) generates output at 162.5 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s).
Latency
0.80s TTFTMistral Small (Sep '24) has a time to first token (TTFT) of 0.80s (based on Mistral's API), which is very competitive compared to other open weight non-reasoning models of similar size (median: 1.54s).
Input Price
$0.20 / 1M tokensMistral Small (Sep '24) costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.60 per 1M output tokens (somewhat higher than average, median: $0.32), based on Mistral's API.
Output Price
$0.60 / 1M tokensMistral Small (Sep '24) costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.60 per 1M output tokens (somewhat higher than average, median: $0.32), based on Mistral's API.

Assessment

Routing verdict and task fit

Mistral Small (Sep '24) scores 5 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 6). Mistral Small (Sep '24) generates output at 162.5 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s). Mistral Small (Sep '24) costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.60 per 1M output tokens (somewhat higher than average, median: $0.32), based on Mistral's API.
Task fit assessment
WorkloadRatingNotes
Complex reasoning & agentic workflowsSuboptimalIntelligence score 5 is better suited for straightforward tasks than multi-step reasoning.
High-volume chat & customer-facingOptimalOutput speed 162.5 tokens/sec is adequate for chat.
Latency-sensitive applicationsOptimalTTFT 0.80s — 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

15 published comparison charts for this profile. Mistral Small (Sep) is marked with a diamond and bold label; each chart has a table view and 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 3.1: 14.7, row 14 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. Qwen3.6 27B30.5
  12. gpt-oss-120b (high)23.8
  13. Devstral Small 217.4
  14. Mistral Small 3.1 (this model)14.7
  15. Ministral 3 14B11.1
  16. Mistral Small 3.2 (this model)10.6
  17. Ministral 3 8B9
  18. Granite 4.1 30B8.9
  19. Llama 3.1 8B7.6
  20. Gemma 3 27B7.4

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 3.1: 14.7, row 14 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. Qwen3.6 27B30.5
  12. gpt-oss-120b (high)23.8
  13. Devstral Small 217.4
  14. Mistral Small 3.1 (this model)14.7
  15. Ministral 3 14B11.1
  16. Mistral Small 3.2 (this model)10.6
  17. Ministral 3 8B9
  18. Granite 4.1 30B8.9
  19. Llama 3.1 8B7.6
  20. Gemma 3 27B7.4

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

12 models in the published order. Mistral Small 3.2: 38.9, row 9 of 12 in the published order.
  1. Nemotron 3 Ultra83.3
  2. Granite 4.1 30B61.1
  3. DeepSeek V4 Pro (max)50
  4. Gemma 3 27B50
  5. Gemma 3 12B50
  6. GLM-5.2 (max)44.4
  7. gpt-oss-120b (high)38.9
  8. Qwen3.6 27B38.9
  9. Mistral Small 3.2 (this model)38.9
  10. Kimi K2.633.3
  11. MiniMax-M333.3
  12. EXAONE 4.5 33B27.8

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 (Sep): 4.7, 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 (Sep) (this model)4.7

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.

Output Speed

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

18 models in the published order. Mistral Small 3.1: 171 t/s, row 5 of 18 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 3.1 (this model)171 t/s
  6. Grok 4.3 (high)164 t/s
  7. Mistral Small (Sep) (this model)163 t/s
  8. Mistral Small 3.2 (this model)147 t/s
  9. Llama 3.1 8B144 t/s
  10. MiniMax-M399 t/s
  11. Ministral 3 8B93 t/s
  12. GPT-5.5 (xhigh)88 t/s
  13. Ministral 3 14B78 t/s
  14. Kimi K2.676 t/s
  15. DeepSeek V4 Pro (max)72 t/s
  16. Claude Fable 5 (with fallback)71 t/s
  17. Devstral Small 268 t/s
  18. Qwen3.6 27B60 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 (Sep): 163 t/s, row 6 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. Grok 4.3 (high)164 t/s
  6. Mistral Small (Sep) (this model)163 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

18 models in the published order. Mistral Small 3.1: 2.92 s, row 1 of 18 in the published order.
  1. Mistral Small 3.1 (this model)2.92 s
  2. Mistral Small (Sep) (this model)3.08 s
  3. Mistral Small 3.2 (this model)3.39 s
  4. Llama 3.1 8B3.48 s
  5. Ministral 3 8B5.35 s
  6. Ministral 3 14B6.4 s
  7. Devstral Small 27.37 s
  8. gpt-oss-120b (high)1.59 s
  9. Qwen3.6 27B8.34 s
  10. Nemotron 3 Ultra2 s
  11. GLM-5.2 (max)2.3 s
  12. Grok 4.3 (high)3.05 s
  13. MiniMax-M35.05 s
  14. Gemini 3.5 Flash2.6 s
  15. Kimi K2.66.56 s
  16. DeepSeek V4 Pro (max)6.97 s
  17. GPT-5.5 (xhigh)5.68 s
  18. Claude Fable 5 (with fallback)7.07 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

18 models in the published order. Mistral Small 3.2: 0 s, row 1 of 18 in the published order.
  1. Mistral Small 3.2 (this model)0 s
  2. Mistral Small 3.1 (this model)0 s
  3. Mistral Small (Sep) (this model)0 s
  4. Ministral 3 8B0 s
  5. Ministral 3 14B0 s
  6. Llama 3.1 8B0 s
  7. Devstral Small 20 s
  8. Qwen3.6 27B0 s
  9. gpt-oss-120b (high)6.37 s
  10. Nemotron 3 Ultra9.09 s
  11. GLM-5.2 (max)9.19 s
  12. Grok 4.3 (high)0 s
  13. MiniMax-M320.19 s
  14. Gemini 3.5 Flash0 s
  15. Kimi K2.658.43 s
  16. DeepSeek V4 Pro (max)60.99 s
  17. GPT-5.5 (xhigh)0 s
  18. Claude Fable 5 (with fallback)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 3.1: 128k, row 18 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. GPT-5.5 (xhigh)922k
  8. Nemotron 3 Ultra262.1k
  9. Muse Spark262.1k
  10. Qwen3.6 27B262.1k
  11. EXAONE 4.5 33B262.1k
  12. Kimi K2.6256k
  13. Devstral Small 2256k
  14. Ministral 3 14B256k
  15. Ministral 3 8B256k
  16. gpt-oss-120b (high)131.1k
  17. Granite 4.1 30B131.1k
  18. Mistral Small 3.1 (this model)128k
  19. Mistral Small 3.2 (this model)128k
  20. Llama 3.1 8B128k

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

15 models in the published order. Mistral Small 3.1: 0.00143, row 1 of 15 in the published order.
  1. Mistral Small 3.1 (this model)0.00143
  2. DeepSeek V4 Pro (max)0.0063
  3. gpt-oss-120b (high)0.00454
  4. Mistral Small 3.2 (this model)0.00242
  5. MiniMax-M30.0139
  6. Grok 4.3 (high)0.0101
  7. Ministral 3 14B0.00235
  8. Ministral 3 8B0.00235
  9. Nemotron 3 Ultra0.0209
  10. Kimi K2.60.0216
  11. Qwen3.6 27B0.041
  12. GLM-5.2 (max)0.0268
  13. Gemini 3.5 Flash0.0884
  14. GPT-5.5 (xhigh)0.107
  15. 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 (Sep) 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

15 models in the published order. Mistral Small 3.1: 1.5, row 1 of 15 in the published order.
  1. Mistral Small 3.1 (this model)1.5
  2. gpt-oss-120b (high)3.5
  3. Ministral 3 8B2.8
  4. Ministral 3 14B3
  5. Mistral Small 3.2 (this model)2.5
  6. DeepSeek V4 Pro (max)7.2
  7. MiniMax-M314.5
  8. Grok 4.3 (high)12.5
  9. Qwen3.6 27B87.9
  10. Nemotron 3 Ultra33.5
  11. GLM-5.2 (max)32.9
  12. Kimi K2.625
  13. Gemini 3.5 Flash114
  14. GPT-5.5 (xhigh)131.8
  15. 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

17 models; segments: Input, Output, Cache Hit. Mistral Small 3.1: $0.400, row 3 of 17 in the published order.
  • Input
  • Output
  • Cache Hit
  1. Llama 3.1 8B$0.170
  2. Ministral 3 8B$0.300
  3. Mistral Small 3.1 (this model)$0.400
  4. Ministral 3 14B$0.400
  5. Mistral Small 3.2 (this model)$0.400
  6. gpt-oss-120b (high)$0.900
  7. Mistral Small (Sep) (this model)$0.800
  8. DeepSeek V4 Pro (max)$1.31
  9. MiniMax-M3$1.56
  10. Nemotron 3 Ultra$3.60
  11. Grok 4.3 (high)$3.95
  12. Qwen3.6 27B$4.20
  13. Kimi K2.6$5.11
  14. GLM-5.2 (max)$6.06
  15. Gemini 3.5 Flash$10.65
  16. GPT-5.5 (xhigh)$35.50
  17. 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

16 models in the published order. Mistral Small 3.1: 0.447, row 1 of 16 in the published order.
  1. Mistral Small 3.1 (this model)0.447
  2. Mistral Small 3.2 (this model)0.87
  3. Nemotron 3 Ultra1.4
  4. Devstral Small 21.4
  5. Grok 4.3 (high)1.6
  6. gpt-oss-120b (high)1.9
  7. Gemini 3.5 Flash2.2
  8. Ministral 3 14B2.2
  9. Ministral 3 8B2.5
  10. GLM-5.2 (max)2.8
  11. Qwen3.6 27B3
  12. GPT-5.5 (xhigh)3
  13. MiniMax-M33.9
  14. Claude Fable 5 (with fallback)4.8
  15. DeepSeek V4 Pro (max)6.8
  16. Kimi K2.68.2

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.

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 3.1: 24B, row 12 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. gpt-oss-120b (high)117B
  7. EXAONE 4.5 33B34.4B
  8. Granite 4.1 30B30B
  9. Qwen3.6 27B27.8B
  10. Gemma 3 27B27.4B
  11. Devstral Small 224B
  12. Mistral Small 3.1 (this model)24B
  13. Mistral Small 3.2 (this model)24B
  14. Mistral Small (Sep) (this model)22B
  15. Ministral 3 14B14B
  16. Gemma 3 12B12.2B
  17. Ministral 3 8B8B
  18. Llama 3.1 8B8B

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.

Specifications

Architecture and limits

Mistral Small (Sep) technical specifications
SpecificationValueSource
Model typeOpen weightsfaq
ReasoningNofaq
Input modalitiesMistral Small (Sep '24) supports text only input.faq
Output modalitiesMistral Small (Sep '24) supports text only output.faq
Context window33k tokensfaq
Open weights / sourceYes, Mistral Small (Sep '24) is open weights. The model weights are publicly available and can be downloaded for self-hosting.faq
ParametersMistral Small (Sep '24) has 22 billion parameters.faq
LicenseMistral Small (Sep '24) is released under the Mistral Research License license. Commercial use requires a separate license agreement.faq
API availabilityYes, Mistral Small (Sep '24) 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 (Sep).

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

Questions

Model FAQ and technical disclosures

Mistral Small (Sep '24) was released on September 17, 2024.

Mistral Small (Sep '24) was created by Mistral.

Mistral Small (Sep '24) scores 5 (estimated) on the Artificial Analysis Intelligence Index, placing it below average among other open weight non-reasoning models of similar size (median: 6).

Mistral Small (Sep '24) generates output at 162.5 tokens per second (based on Mistral's API), which is well above average compared to other open weight non-reasoning models of similar size (median: 101.8 t/s).

Mistral Small (Sep '24) has a time to first token (TTFT) of 0.80s (based on Mistral's API), which is very competitive compared to other open weight non-reasoning models of similar size (median: 1.54s).

Mistral Small (Sep '24) costs $0.20 per 1M input tokens (somewhat higher than average, median: $0.15) and $0.60 per 1M output tokens (somewhat higher than average, median: $0.32), based on Mistral's API.

Mistral Small (Sep '24) costs $0.20 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 $0.24 per 1M tokens. Pricing may vary by provider.

No, Mistral Small (Sep '24) is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.

Mistral Small (Sep '24) supports text only input.

Mistral Small (Sep '24) supports text only output.

No, Mistral Small (Sep '24) does not support image input. It can only process text.

No, Mistral Small (Sep '24) is not multimodal. It only supports text only input.

Mistral Small (Sep '24) has a context window of 33k tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Mistral Small (Sep '24) is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Mistral Small (Sep '24) has 22 billion parameters.

Mistral Small (Sep '24) is released under the Mistral Research License license. Commercial use requires a separate license agreement.

Mistral Small (Sep '24) achieves a score of 5 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.

Yes, Mistral Small (Sep '24) is available via API through 1 provider.

Mistral Small (Sep '24) is available through 1 API provider.

Models

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