Mistral Large Instruct 2411
Mistral AI
Multilingual Tool Calls
Mistral Large Instruct 2411 is a 122.61-billion-parameter dense transformer from Mistral AI, optimized for advanced reasoning, multilingual conversation, and native function calling. It extends the earlier Mistral Large 2407 with improved long-context handling, system prompt adherence, and agentic capabilities. The model supports tool calling and 10 languages including English, French, German, and Chinese, with training on over 80 programming languages. With a 128K context window and flash attention, it handles document-scale analysis and can be self-hosted with GGUF quantization for local inference.
Hardware Configuration
Optional — for precise deployment recommendations
| Quantization | Quality | Size | Fit |
|---|---|---|---|
| Q8_0 | High | 121.33 GB | — |
| Q6_K | High | 93.67 GB | — |
| Q5_K_M | Medium | 80.55 GB | — |
| Q5_K_S | Medium | 78.56 GB | — |
| Q4_K_M | Medium | 68.19 GB | — |
| Q4_K_S | Medium | 64.79 GB | — |
| Q4_0 | Medium | 64.57 GB | — |
| Q3_K_M | Low | 55.04 GB | — |
| Q3_K_S | Low | 49.22 GB | — |
| Q3_K_XL | Low | 60.45 GB | — |
| Q2_K | Low | 42.09 GB | — |
| Q2_K_L | Low | 42.46 GB | — |
| Q3_K_L | Low | 60.12 GB | — |
| Q4_0_8_8 | Low | 64.33 GB | — |
Last updated: March 5, 2026