Mistral Small 3.2

By Mistral · released 2025-06-20 (14 months) · snapshot 2026-09-09

Off the frontierNot on frontier at launch dayopen weights
Intelligence index
7.00
Cost to run the AA index
$163
Price per 1M tokens (in / out)
$0.10 / $0.30
End-to-end time
3.90s
Better than…

Is Mistral Small 3.2 on the Pareto frontier, and what beats it?

Off the frontier

No. MiMo-V2.5-Pro is at least as smart for the same money or less.

Next step up: GPT-5.6 Luna (xhigh), smarter but dearer.

Mistral Small 3.2 vs MiMo-V2.5-ProMistral Small 3.2 vs GPT-5.6 Luna (xhigh)
#ModelCreatorIntelligenceTotal costE2E totalOpenReleasedFrontier
1Mistral Small 3.2Mistral7.00$1633.90s2025-06-20
2MiMo-V2.5-ProXiaomi26.40$13872.6s2026-04-22Frontier
3GPT-5.6 Luna (xhigh)OpenAI34.80$17960.5s2026-07-09Frontier

Open in the picker Daily check URL: 200 on the frontier, 409 once beaten

At a glance

Mistral Small 3.2 today

Time to first answer0.72s
Output speed157 tok/s
Input modalitiesNot reported
0102030405060$20$50$100$200$500$1k$2k$5k$10k$20kCost to run the Artificial Analysis Intelligence Index (USD, log scale)Intelligence IndexCeleris-1 (Celeris): intelligence 6.30, $60.0Claude 4.5 Haiku (Reasoning) (Anthropic): intelligence 17.60, $524Claude 4.5 Sonnet (Reasoning) (Anthropic): intelligence 21.20, $1501Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) (Anthropic): intelligence 49.70, $11161Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback) (Anthropic): intelligence 51.20, $5242Claude Fable 5.1 (Adaptive Reasoning, Low Effort, Default Fallback) (Anthropic): intelligence 47.00, $3158Claude Fable 5.1 (Adaptive Reasoning, Medium Effort, Default Fallback) (Anthropic): intelligence 49.10, $3983Claude Opus 4.8 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 42.00, $6874Claude Opus 5 (Adaptive Reasoning, High Effort) (Anthropic): intelligence 48.20, $4332Claude Opus 5 (Adaptive Reasoning, Low Effort) (Anthropic): intelligence 39.80, $1561Claude Opus 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 50.70, $7275Claude Opus 5 (Adaptive Reasoning, Medium Effort) (Anthropic): intelligence 45.10, $2732Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) (Anthropic): intelligence 49.70, $5868Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 30.50, $4925Claude Sonnet 5 (Adaptive Reasoning, Low Effort) (Anthropic): intelligence 24.70, $653Claude Sonnet 5 (Adaptive Reasoning, Max Effort) (Anthropic): intelligence 38.40, $6998Claude Sonnet 5 (Adaptive Reasoning, Medium Effort) (Anthropic): intelligence 28.40, $1218DeepSeek R1 (Jan '25) (DeepSeek): intelligence 11.40, $461DeepSeek V3 (Dec '24) (DeepSeek): intelligence 8.50, $25.4DeepSeek V3 0324 (DeepSeek): intelligence 9.70, $87.2DeepSeek V4 Flash (Reasoning, High Effort) (DeepSeek): intelligence 24.80, $210DeepSeek V4 Flash (Reasoning, Max Effort) (DeepSeek): intelligence 24.60, $191DeepSeek V4 Flash 0731 (Reasoning, Max Effort) (DeepSeek): intelligence 34.50, $474DeepSeek V4 Flash Vision (Reasoning, Max Effort) (DeepSeek): intelligence 35.00, $445DeepSeek V4 Pro (Reasoning, Max Effort) (DeepSeek): intelligence 30.90, $277DeepSeek V4 Pro 0813 (Reasoning, Max Effort) (DeepSeek): intelligence 36.30, $1122Gemini 2.5 Pro (Google): intelligence 16.70, $707Gemini 3.1 Flash-Lite (Google): intelligence 16.00, $119Gemini 3.1 Pro Preview (Google): intelligence 30.40, $1310Gemini 3.5 Flash (high) (Google): intelligence 33.00, $2172Gemini 3.5 Flash-Lite (Google): intelligence 22.70, $267Gemini 3.6 Flash (high) (Google): intelligence 34.30, $1037Gemini 3.7 Flash (high) (Google): intelligence 39.40, $1084Gemini 3.8 Flash (high) (Google): intelligence 41.20, $1623Gemini 3.8 Flash (medium) (Google): intelligence 40.00, $1100GLM-5.3 (max) (Z AI): intelligence 44.90, $2503GPT-5 mini (high) (OpenAI): intelligence 17.40, $189GPT-5.4 mini (xhigh) (OpenAI): intelligence 24.60, $1438GPT-5.4 nano (xhigh) (OpenAI): intelligence 21.20, $331GPT-5.5 (high) (OpenAI): intelligence 37.30, $3411GPT-5.5 (medium) (OpenAI): intelligence 34.20, $2095GPT-5.5 (xhigh) (OpenAI): intelligence 38.60, $5294GPT-5.5 Instant (June 2026) (OpenAI): intelligence 26.80, $1184GPT-5.6 Luna (max) (OpenAI): intelligence 37.50, $320GPT-5.6 Sol (low) (OpenAI): intelligence 33.80, $637GPT-5.6 Sol (max) (OpenAI): intelligence 47.10, $3465GPT-5.6 Sol (medium) (OpenAI): intelligence 39.50, $997GPT-5.6 Sol (xhigh) (OpenAI): intelligence 44.10, $2082GPT-5.6 Terra (high) (OpenAI): intelligence 34.50, $768GPT-5.6 Terra (max) (OpenAI): intelligence 42.30, $2501GPT-5.6 Terra (xhigh) (OpenAI): intelligence 38.20, $1187GPT-6 Astra (Non-reasoning) (OpenAI): intelligence 45.20, $2441gpt-oss-120b (high) (OpenAI): intelligence 12.30, $122gpt-oss-20b (high) (OpenAI): intelligence 9.00, $29.4Grok 4.3 (high) (SpaceXAI): intelligence 25.40, $332Grok 4.3 (Non-reasoning) (SpaceXAI): intelligence 14.50, $203Grok 4.5 (high) (SpaceXAI): intelligence 39.10, $1404Grok 4.6 (high) (SpaceXAI): intelligence 44.40, $2352Grok 4.6 (low) (SpaceXAI): intelligence 35.40, $761Grok 4.6 (medium) (SpaceXAI): intelligence 43.00, $1937Grok 4.6 (xhigh) (SpaceXAI): intelligence 44.30, $2830Hy3 (Tencent): intelligence 25.80, $149Inkling (xhigh) (Thinking Machines): intelligence 25.50, $1026Inkling Small (Thinking Machines): intelligence 26.10, $241Kimi K2.7 Code (Kimi): intelligence 26.30, $993Kimi K3 (max) (Kimi): intelligence 43.80, $3658Llama 4 Maverick (Meta): intelligence 9.30, $45.5Llama 4 Scout (Meta): intelligence 6.50, $77.6LongCat 2.0 (LongCat): intelligence 19.70, $545Mercury 2 (Inception): intelligence 11.50, $153MiniMax-M2.7 (MiniMax): intelligence 23.20, $236MiniMax-M3 (MiniMax): intelligence 29.60, $538Ministral 3 14B (Mistral): intelligence 6.00, $250Ministral 3 3B (Mistral): intelligence 4.80, $61.3Ministral 3 8B (Mistral): intelligence 5.50, $169Mistral Large 3 (Mistral): intelligence 9.70, $130Mistral Medium 3.5 (Mistral): intelligence 14.90, $1160Mistral Small 3.1 (Mistral): intelligence 7.40, $35.8Mistral Small 4 (Reasoning) (Mistral): intelligence 11.50, $78.7Muse Glimmer (high) (Meta): intelligence 18.10, $152Muse Spark 1.1 (xhigh) (Meta): intelligence 34.30, $1737Muse Spark 1.2 (xhigh) (Meta): intelligence 39.80, $1385Muse Spark 1.3 (xhigh) (Meta): intelligence 45.20, $1655Nemotron 3 Super 120B A12B (Reasoning) (NVIDIA): intelligence 13.60, $859Nemotron 3 Ultra 550B A55B (Reasoning) (NVIDIA): intelligence 23.40, $801Nemotron 3.5 Lightning (NVIDIA): intelligence 13.60, $105NVIDIA Nemotron 3 Nano 30B A3B (Reasoning) (NVIDIA): intelligence 8.90, $63.7Quasar 438B (max, based on GLM-5.2) (Multiverse Computing): intelligence 27.10, $2504Qwen3 235B A22B 2507 (Reasoning) (Alibaba): intelligence 12.70, $235Qwen3 30B A3B 2507 (Reasoning) (Alibaba): intelligence 9.80, $188Qwen3 Coder Next (Alibaba): intelligence 10.10, $476Qwen3.5 122B A10B (Reasoning) (Alibaba): intelligence 16.20, $661Qwen3.5 397B A17B (Reasoning) (Alibaba): intelligence 19.10, $941Qwen3.6 27B (Reasoning) (Alibaba): intelligence 21.90, $1145Qwen3.7 Max (Alibaba): intelligence 29.90, $2277Qwen3.7 Plus (Alibaba): intelligence 25.80, $568Qwen3.8 2.4T A95B (Alibaba): intelligence 40.00, $2632Qwen3.8 27B (xhigh) (Alibaba): intelligence 33.90, $1170Qwen3.8 Max (Alibaba): intelligence 40.30, $3172Ring-2.6-1T (InclusionAI): intelligence 17.30, $576Solar Pro 3 (Upstage): intelligence 7.80, $159Trinity Large Thinking (Arcee AI): intelligence 10.90, $518Granite 4.2 3B (IBM): intelligence 9.10, $14.9MiMo-V2.5 (Xiaomi): intelligence 22.30, $42.6MiMo-V2.5-Pro (Xiaomi): intelligence 26.40, $138GPT-5.6 Luna (xhigh) (OpenAI): intelligence 34.80, $179GLM-5.3-Flash (Z AI): intelligence 41.90, $280GPT-5.6 Sol (high) (OpenAI): intelligence 42.50, $1487GPT-6 Astra (low) (OpenAI): intelligence 46.00, $1537Muse Spark 1.3 (max) (Meta): intelligence 48.20, $2000GPT-6 Astra (medium) (OpenAI): intelligence 49.70, $2434GPT-6 Astra (high) (OpenAI): intelligence 51.00, $2917GPT-6 Astra (xhigh) (OpenAI): intelligence 52.50, $3803GPT-6 Astra (max) (OpenAI): intelligence 52.80, $5324Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback) (Anthropic): intelligence 53.20, $9063Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback) (Anthropic): intelligence 53.40, $13129Mistral Small 3.2 (Mistral): intelligence 7.00, $163Mistral Small 3.2Granite 4.2 3BMiMo-V2.5-ProGPT-5.6 Luna (xhigh)Claude Fable 5.1 (max)
Mistral Small 3.2Same release, other effort levelsToday's Pareto frontierOpen-weights frontierEvery other model
Up and left is better. Dotted line: today's Pareto frontier. Hover a point for its numbers, click it for its page.

Intelligence: Artificial Analysis Intelligence Index. Cost: total cost to run that benchmark, so reasoning tokens count. API id: mistral-small-3-2.

Matchup

Compare head to head

Suggested: what beats it, its effort levels, and frontier models near its price.

Compare Mistral Small 3.2 with measured models

Release

What the Mistral Small 3.2 release did to the frontier on 2025-06-20

None of its 1 effort levels reached the frontier at launch.

Closest miss: Mistral Small 3.2, 1.5 points below DeepSeek V3 (Dec '24) at 6.4× the cost.

Release page: launch-day frontier and every effort level

Deciding

Should you use Mistral Small 3.2?

Should you use Mistral Small 3.2?

Only when it fits your setup: MiMo-V2.5-Pro is at least as smart for the same money or less, so Mistral Small 3.2 is off the value frontier today.

What is better and cheaper than Mistral Small 3.2?

MiMo-V2.5-Pro — at least as smart for the same money or less.

How much does Mistral Small 3.2 cost?

$0.10/$0.30 in/out per 1M tokens, and $163 to run the full Artificial Analysis Intelligence Index, the honest all-in number for reasoning models.

Is Mistral Small 3.2 on the Pareto frontier?

No: MiMo-V2.5-Pro beats it on intelligence, cost, or both.

Frontier membership by axis pair

Other axis pairs where Mistral Small 3.2 is Pareto-optimal.

Y axisX axisOn the frontier
AA intelligence indexAA total cost to run (USD)
AA intelligence indexEnd-to-end total seconds
AA intelligence indexOutput speed tokens/sec
Agents: Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/models/mistral-small-3-2 · Markdown: /api/v1/models/mistral-small-3-2?format=md · llms.txt · agent skill