Mistral Large 4 1T-A52B
Mistral Large 4 is a roughly 1T-total, 52B-active multimodal MoE for instruction following, reasoning, coding, agents, function calling, and image understanding. It supports a 1M-token context and ships as an official FP8 checkpoint under the Mistral Ecosystem Open-Weight License. The base weights are server class; an official NVFP4 checkpoint targets newer Blackwell GPUs.
Task Fit
Not marked for typed decisions in the current library.
Tool use, repo work, terminal workflows, and coding benchmarks.
Code generation, debugging, refactoring, and benchmark signal.
General writing, Q&A, and assistant use.
Document QA benefits from long context and instruction following.
Image or visual understanding, not necessarily image generation.
Not marked for image generation in the current library.
Not marked for video generation in the current library.
Not marked for voice in the current library.
Evidence level
How this is assessed →Memory and hardware-fit values are planning estimates unless a tested report is linked.
Variants and Quant Artifacts
Choose the artifact first; hardware fit follows from RAM, VRAM, format, and runtime.
| Quant | Format | Quality | Min RAM | Reco RAM | Runtime | Action |
|---|---|---|---|---|---|---|
| FP8 | safetensors | balanced | 1TB | 1TB | transformers, vllm, sglang, llama.cpp, docker-model-runner | Plan with this |
Recommended Hardware
No compatible local hardware found in the current hardware pool.
Benchmarks
No benchmark data is available for this model yet.
Sources and compatibility
Execution evidence
Run Mistral Large 4 1T-A52B with a documented recipe
Recipes connect hardware, a model artifact, tools, settings, verification, and a reportable result.
No verified recipe is linked to this record yet.
Compatibility estimates remain available in the planner. A recipe appears here only after its exact stack and verification protocol are documented.
Similar Models
DeepSeek V4.1 Flash is DeepSeek's MIT-licensed open-weight multimodal MoE for reasoning, coding, agents, and native image understanding. The complete checkpoint is approximately 763B parameters, including a 552B backbone; its Causal Encoder-Decoder architecture activates 8B parameters during input prefill and 16B during output decoding. CSA2 and FP4 KV caching reduce long-context cache cost. The model supports a 1M-token context window and is available through official FP8 weights, the DeepSeek API, and Ollama Cloud. It remains a data-center-class deployment despite the low active-parameter count.
DeepSeek V4.1 Flash 1017 is an updated MIT-licensed 763B-parameter multimodal MoE checkpoint. It keeps the V4.1 family's 1M-token context, coding, reasoning, tool use, and image understanding, while remaining a data-center-class deployment despite activating far fewer parameters per token.
Kimi K3.1 Pro is Moonshot AI's open-weight, natively multimodal 2.8T-parameter MoE with 104B active parameters and a 1M-token context window. The official checkpoint uses MXFP4 weights with MXFP8 activations and targets coding, research, long-running agents, tool use, and image understanding. Its roughly 1.56 TB repository remains a distributed server-class deployment.