Mistral Medium 3 vs Mistral Small 3
Benchmarks, API Token Pricing, Context Window & Speed Comparison
Related topics (12)
Mistral AI 的 Mistral Medium 3 与 Mistral AI 的 Mistral Small 3 是当前备受关注的两款主力 AI 模型。Mistral Medium 3 提供 131.1K Token 上下文窗口，API 定价为每百万 Token 输入 $0.40 / 输出 $2.00；而 Mistral Small 3 提供 32K Token 上下文窗口，定价为每百万 Token 输入 $0.10 / 输出 $0.30。以下为两者的基准测试、延迟与成本全面对比。
切换并对比任意模型
Select any AI models from our verified benchmark database to generate live side-by-side matrices.
Mistral Medium 3
开源权重基础大模型Mistral AI 推出的 Mistral Medium 3 是一款支持私有云服务器自托管、微调与完全数据主权的开源权重基础大模型，具备 131K 上下文窗口、85 Token/秒生成速度与 47/100 综合智能指数（每百万 Token $0.4 输入 / $2 输出）。
$0.40 / 1M
$2.00 / 1M
131.1K tokens
40.5%
Mistral Small 3
开源权重基础大模型Mistral AI 推出的 Mistral Small 3 是一款支持私有云服务器自托管、微调与完全数据主权的开源权重基础大模型，具备 32K 上下文窗口、105 Token/秒生成速度与 41/100 综合智能指数（每百万 Token $0.1 输入 / $0.3 输出）。
$0.10 / 1M
$0.30 / 1M
32K tokens
31.5%
核心差异一览
- Mistral Small 3 is 5.2x cheaper for API usage (input + output tokens combined).
- Mistral Medium 3 has a larger context window (131K vs 32K tokens).
- Mistral Small 3 is faster at 105 tok/s vs 85 tok/s.
- Mistral Medium 3 leads on SWE-bench coding (40.5% vs 31.5%).
| Feature | Mistral Medium 3 | Mistral Small 3 |
|---|---|---|
Intelligence Index Composite reasoning & knowledge score (0-100) | 47 | 41 |
Provider | Mistral AI | Mistral AI |
Input Price Cost per million input tokens | $0.40 / 1M | $0.10 / 1M |
Output Price Cost per million output tokens | $2.00 / 1M | $0.30 / 1M |
Blended Price Weighted 7:2:1 cache:input:output per 1M | $0.35 / 1M | $0.09 / 1M |
Context Window Maximum input + output tokens | 131.1K tokens~197 A4 pages of text | 32K tokens~48 A4 pages of text |
Max Output Tokens Per single request | 16K | 8K |
Throughput Speed Tokens generated per second | 85 tok/s | 105 tok/s |
SWE-bench Verified Real-world software engineering | 40.5% | 31.5% |
GPQA Diamond Graduate-level reasoning | 57.5% | 46.5% |
Cached Input Price Prompt caching discount per 1M tokens | $0.10 / 1M | $0.05 / 1M |
各项指标谁更胜一筹？
At-a-glance decision framework权威基准测试得分对比
Official evaluation metrics verified by independent third-party research labs.
Industry benchmark evaluating real-world GitHub issue resolution and multi-file debugging.
Challenging questions created by domain experts to test complex PhD-level reasoning.
Hardened reasoning and academic knowledge benchmark with reduced lucky-guess probability.
Crowdsourced human preference evaluation based on blind, head-to-head model battles.
Autonomous multi-step tasks executed inside a real command-line terminal environment.
American Invitational Mathematics Examination — multi-step Olympiad-level reasoning.
Accuracy when reasoning over very large inputs that fill the model context window.
Tokens emitted per second during active generation streaming.
延迟与响应速度
Time to first token (TTFT) and estimated end-to-end response time for a typical 500-token response.
实时 Token 成本模拟器
Adjust sliders to calculate projected monthly cloud API spending with prompt caching discounts.
Mistral Medium 3 · Strengths & Key Advantages
- 提供 131K 上下文窗口与最高 16K 输出容量，核心基准表现优异（SWE-bench Verified 40.5% · GPQA Diamond 57.5%）
- 高达 85 Token/秒的生成吞吐量，首字延迟 (TTFT) 仅约 480 毫秒
- 开放模型权重 (Confidential (~70B))，支持通过 vLLM、SGLang 或 Ollama 在私有 GPU 集群上本地部署与微调
Considerations & Trade-offs
- 在本地以满血精度运行需要较高的 GPU 显存 (VRAM) 与运维基础设施
- 在超长链路的前沿科研推导任务上，深度略逊于顶配重型旗舰推理模型
Mistral Small 3 · Strengths & Key Advantages
- 提供 32K 上下文窗口与最高 8K 输出容量，核心基准表现优异（SWE-bench Verified 31.5% · GPQA Diamond 46.5%）
- 高达 105 Token/秒的生成吞吐量，首字延迟 (TTFT) 仅约 380 毫秒
- 开放模型权重 (24B)，支持通过 vLLM、SGLang 或 Ollama 在私有 GPU 集群上本地部署与微调
Considerations & Trade-offs
- 在本地以满血精度运行需要较高的 GPU 显存 (VRAM) 与运维基础设施
- 在超长链路的前沿科研推导任务上，深度略逊于顶配重型旗舰推理模型
Token Use & Task Economics
Estimated cost and token consumption to complete one representative agentic task.
| Metric | Mistral Medium 3 | Mistral Small 3 |
|---|---|---|
Cost per Task Estimated USD to finish one task | $0.35 | $0.08 |
Output Tokens / Task Tokens generated per task | 35K | 25K |
Reasoning Tokens / Task Thinking tokens burned per task | 0K | 0K |
技术规格参数
Architecture, release timeline, capabilities, and modalities.
| Feature | Mistral Medium 3 | Mistral Small 3 |
|---|---|---|
| Official API Model ID (SDK) | mistral-medium-3-latest | mistral-small-3-latest |
| API Protocol | OpenAI-Compatible REST / vLLM / SGLang / HuggingFace TGI | OpenAI-Compatible REST / vLLM / SGLang / HuggingFace TGI |
| Release Date | 2025-05 | 2025-01 |
| Knowledge Cutoff | March 2025 | December 2024 |
| Architecture | Enterprise Multimodal Transformer | Compact Dense Transformer |
| Parameters | Confidential (~70B) | 24B |
| Enterprise Rate Limits (RPM / TPM) | Unlimited (Hardware-bound on self-hosted GPU) / 10,000 RPM Cloud · Unlimited local throughput / 10M+ TPM Managed Cloud | Unlimited (Hardware-bound on self-hosted GPU) / 10,000 RPM Cloud · Unlimited local throughput / 10M+ TPM Managed Cloud |
| Long-Context Tier (>128K/200K) | Flat rate ($0.4/1M in) across full 131K window | Flat rate ($0.1/1M in) across full 32K window |
| Prompt Cache Policy & TTL | Cached input at $0.1/1M tokens | Cached input at $0.05/1M tokens |
| Computer Use / GUI Agent | API / CLI Tool-Use & DOM Agent ready (no raw pixel desktop click loop) | API / CLI Tool-Use & DOM Agent ready (no raw pixel desktop click loop) |
| MCP & Parallel Tool Calling | Native MCP Client/Server ready · Parallel function calling · 100% Strict JSON Schema | Native MCP Client/Server ready · Parallel function calling · 100% Strict JSON Schema |
| Web Search Grounding & Sandbox | Supported via external Tavily/Exa Search & E2B/Docker Sandbox tools | Supported via external Tavily/Exa Search & E2B/Docker Sandbox tools |
| Official Fine-Tuning Support | Full Weights SFT, LoRA, QLoRA, DPO & GRPO RL (Unsloth / Axolotl / Llama-Factory) | Full Weights SFT, LoRA, QLoRA, DPO & GRPO RL (Unsloth / Axolotl / Llama-Factory) |
| Multimodal Token Billing | Text BPE (~4 chars/token) · Vision (~258–765 tokens per image tile) | Text BPE (~4 chars/token) |
| Quantization & Min GPU VRAM | BF16 / FP16, FP8 Dynamic, INT4 AWQ / GPTQ, GGUF (Q4_K_M / Q8_0) · 43 GB VRAM (4-bit Quantized) · 151 GB VRAM (Native BF16) | BF16 / FP16, FP8 Dynamic, INT4 AWQ / GPTQ, GGUF (Q4_K_M / Q8_0) · 15 GB VRAM (4-bit Quantized) · 52 GB VRAM (Native BF16) |
| Safety & Frontier Classification | Open-Weights System Card + Customizable Guardrail Layer (Llama Guard / ShieldGemma) | Open-Weights System Card + Customizable Guardrail Layer (Llama Guard / ShieldGemma) |
| Enterprise Compliance (ZDR/HIPAA/SOC2) | 100% Air-Gapped On-Premises / Private VPC · Full HIPAA, GDPR & SOC2 Control | 100% Air-Gapped On-Premises / Private VPC · Full HIPAA, GDPR & SOC2 Control |
| Input Modalities | text, vision | text |
| Output Modalities | text | text |
Mistral Medium 3 vs Mistral Small 3: 常见问题解答 (FAQ)
Which is more intelligent, Mistral Medium 3 or Mistral Small 3?
Mistral Medium 3 is more intelligent. It scores 47 compared with 41 on the composite Intelligence Index.
Which model is faster?
Mistral Small 3 is faster, generating 105 tokens per second compared with 85 tokens per second.
Which model is cheaper to run?
Mistral Small 3 is cheaper at $0.09 per 1M tokens blended, compared with $0.35 (7:2:1 cache:input:output ratio).
Which model has lower latency?
Mistral Small 3 has lower latency, with a time to first token of 0.38s compared with 0.48s.
Which model has a larger context window?
Mistral Medium 3 has a larger context window, supporting 131.1K tokens compared with 32K tokens.
综合结论
综合来看，Mistral Medium 3（适合：支持私有云服务器自托管、微调与完全数据主权的开源权重基础任务、企业级 RAG 检索增强生成、自动化智能体工作流与高可靠 API 集成。）与 Mistral Small 3（适合：支持私有云服务器自托管、微调与完全数据主权的开源权重基础任务、企业级 RAG 检索增强生成、自动化智能体工作流与高可靠 API 集成。）各具优势：Mistral Medium 3 在综合智能与编程基准测试中表现更强，Mistral Small 3 拥有更快的 Token 生成速度，而 Mistral Small 3 则在规模化 API 部署中具备更高的成本效益。
常见问题 (FAQ)
在编程和深度推理方面，Mistral Medium 3 和 Mistral Small 3 哪个更强？+
在综合智能指数与编程基准评测中，Mistral Medium 3 表现占优。建议结合上方的 SWE-bench Verified 与 GPQA Diamond 跑分根据具体场景选择。
Mistral Medium 3 与 Mistral Small 3 的 API 价格哪个更便宜？+
Mistral Medium 3 的 API 价格为输入 $0.40 / 输出 $2.00（每百万 Token），Mistral Small 3 为输入 $0.10 / 输出 $0.30。综合混合成本来看，Mistral Small 3 更具性价比。
哪款模型拥有更大的上下文窗口和更快的响应速度？+
Mistral Medium 3 支持 131.1K Token 上下文窗口，而 Mistral Small 3 支持 32K Token。在流式生成速度方面，Mistral Small 3 的吞吐表现更快。
Understand key metrics like context window, token pricing, reasoning mode, and SWE-bench scores in our comprehensive AI glossary.
Related reading
- Aya Vision 8B vs Mistral Small 3：基准测试、API 价格与速度全面对比 (2026)对比
- Command R7B Arabic vs Mistral Small 3：基准测试、API 价格与速度全面对比 (2026)对比
- Llama 3.1 Nemotron 70B Instruct vs Mistral Medium 3：基准测试、API 价格与速度全面对比 (2026)对比
- Mistral Small 3.2 vs Mistral Medium 3：基准测试、API 价格与速度全面对比 (2026)对比
- Aion-3.0 vs Aion-Mini：基准测试、API 价格与速度全面对比 (2026)对比
- Aion-3.0 vs Gemini 3.6 Flash：基准测试、API 价格与速度全面对比 (2026)对比
Benchmark scores, token prices, and rate limits evolve rapidly as AI labs release model updates. Always verify current specifications and pricing on the official provider documentation.
数据来源: Mistral AI 与 Mistral AI 官方 API 文档、模型发布说明及独立基准测试报告 (2026)。