Open-Weight Mistral AI · ministral-8b-2512 · Text+Image+PDF
Ministral 3 8B Official Specifications, Pricing & Benchmarks
Ministral 3 8B (ministral-8b-2512) by Mistral AI, released 2025-12-02, featuring 262,144 context window, 262,144 max output tokens, and Text+Image+PDF input.
📋 Ministral 3 8B Official API & System Card Specifications
Verified against official Mistral AI developer documentation & pricing
1. Identity & Capabilities
API Spec- API Model ID
- ministral-8b-2512
- Release Date
- 2025-12-02
- Weight License
- Open-Weight (API Spec)
- Native Reasoning
- —
- Function Calling & JSON Schema
- ✅ 原生支持
- Batch API 异步批处理
- ✅ Supported (50% off)
2. Context, Output & Modalities
Limits- Max Context Window
- 262,000 tokens (262K)
- Max Output Tokens
- 262,144 tokens
- Input Modalities
- Text + Image + PDF
- Output Modalities
- Text
- Direct China Access
- 🌐 Global / OpenRouter
- Deployment Mode
- Open-Weight + Cloud API
3. Official Pricing & Cache
USD / 1M- Input / 1M Tokens
- $ 0.15
- Cached Input / 1M
- $ 0.015 -90%
- Output / 1M Tokens
- $ 0.15
🎛️ Official Sampling & Reasoning Parameters
Recommended generation settings & API parameter constraints
Standard Generation Config
temperature=0.7
top_p=0.95
max_tokens=65,536
Closest Self-Hostable Open-Weight Alternatives to Ministral 3 8B
Open-weight models with full HuggingFace weights for private GPU / Mac deployment
GLM-5 (744B-A40B MoE)
78L · MIT Q4 需 461.99 GB
DeepSeek-V4-Pro (1.6T MoE)
61L · MIT Q4 需 990.02 GB
Kimi K2.6 (1T MoE)
61L · Modified MIT Q4 需 620.83 GB
Qwen3.5-397B-A17B (MoE)
60L · Apache-2.0 Q4 需 247.5 GB
Kimi K3 (2.8T MoE)
93L · Kimi-K3 Open License Q4 需 1733.2 GB
Qwen3.5-122B-A10B (MoE)
48L · Apache-2.0 Q4 需 77.26 GB
Frequently Asked Questions
Is Ministral 3 8B open-source or proprietary?
Ministral 3 8B (ministral-8b-2512) is a proprietary commercial model by Mistral AI. Weights and internal layer dimensions are undisclosed; it is accessed via cloud API.
What are the context limit and API pricing of Ministral 3 8B?
Ministral 3 8B supports 262K input context and 262,144 max output tokens, priced at $0.15/1M input ($0.015/1M cached) and $0.15/1M output.