Mistral
Devstral Small 2 24B Instruct
前沿
236.7K下载量606点赞Nov 2025发布日期256K tokens上下文Mistral Research License许可证96 卓越质量
Devstral Small 2 24B Instruct (24B parameters) requires approximately 18.9 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 22 GB of VRAM.
快速开始
— 复制粘贴即可本地运行Copy-paste commands to run Devstral Small 2 24B Instruct on your machine.
Run
ollama run devstral-small-2Quick specs
Parameters24B
Architecturedense
Context256K tokens
Modalitytext
Min RAM9.4 GB
Rec. RAM14.6 GB (Q4_K_M)
LicenseMistral Research License
FamilyDevstral Small
✓ Code✓ Reasoning
About this model
- •Agentic Coding: Devstral is designed to excel at agentic coding tasks, making it a great choice for software engineering agents
- •Lightweight: with its compact size of just 24 billion parameters, Devstral is light enough to run on a single RTX 4090 or a Mac with 32GB RAM,...
- •Apache 2.0 License: Open-source license allowing usage and modification for both commercial and non-commercial purposes
- •Context Window: A 256k context window
- •Vision Capabilities: Enables the model to analyze images and provide insights based on visual content, in addition to text
相关模型
快速推荐
最佳硬件
Devstral Small 2 24B Instruct 的最佳选择
运行此模型
量化选项
各量化级别的 VRAM 估算
No hardware detected — fit column shows raw VRAM estimates
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 9.4 GB | Low | — |
Q3_K_S | 3 | 11.8 GB | Low | — |
NVFP4 | 4 | 13.4 GB | Medium | — |
Q4_K_M | 4 | 14.6 GB | Medium | — |
Q5_K_M | 5 | 17.3 GB | High | — |
Q6_K | 6 | 19.7 GB | High | — |
Q8_0 | 8 | 25.7 GB | Very High | — |
F16 | 16 | 49.2 GB | Maximum | — |
Quality benchmarks
Devstral Small 2 24B Instruct benchmark scores
Coding
SWE-bench Verified68.0%
HumanEval+—
Aider Polyglot—
LiveCodeBench—
Source: official · 2025-12-15
硬件兼容性
全部硬件的适配估算
Computing compatibility...
内存详细分析
Reference: RTX 2060 6GB
Weights14.6 GB
KV Cache2.4 GB
Runtime1.2 GB
Headroom0.6 GB
常见问题
FAQ — Devstral Small 2 24B Instruct
另请参阅