Unverified50% confidenceFactExact time
Mistral Large 4 supports a 1-million-token context window and is offered in FP8 and NVFP4 versions.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/31/2026
First Seen
Valid until: 1/29/2027
Sources
Related Entities
Related Claims
UnverifiedMistral Large 4 offers an NVFP4 quantized version based on NVIDIA's Blackwell architecture, which the company calls lossless quantization.76% similarUnverifiedMistral Large 4 uses a 72-layer decoder-only architecture, paired with an image encoder to achieve full-modality support.74% similarUnverifiedOn the scientific coding Sycode benchmark, Mistral Large 4 ranks first among open-weight models.72% similarUnverifiedIt is officially recommended that using the NVFP4 version of Mistral Large 4 requires at least 8 H100 or H200 GPUs.72% similarUnverifiedMistral Large 4 is equipped with a 1.6-billion-parameter vision encoder and has multimodal capabilities.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1120226API
curl https://kongchang.com/api/v1/knowledge/claims/1120226MCP
get_claim(id=1120226)