Unverified50% confidenceFactExact time
Mistral Large 3 supports a native context window of 512,000 tokens, combining block sparse attention with hardware-calibrated rotary position embeddings (RoPE).
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMistral Large 4 natively supports multimodality, with a focus on coding and agentic tasks, and full weights will be available for download at the end of this month.66% similarUnverifiedMistral has launched the Large 4 model, whose core positioning is to power general-purpose agents.66% similarPartially VerifiedMistral Large 4 was trained using 3,800 NVIDIA Grace Blackwell GPUs in its own data centers, and the public preview also runs on the same infrastructure.62% similarUnverifiedOpen-source models with fewer than 7B parameters, such as Llama 3 and Mistral 7B, combined with 4-bit quantization, can run on a single consumer-grade GPU or a high-performance CPU.60% similarUnverifiedOn the CyberGym benchmark, Mistral Large 4 scored 82, ranking first among all open-source models and surpassing MIMO, Grok, GLM 5.3, Kimi K3, and others.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1008181API
curl https://kongchang.com/api/v1/knowledge/claims/1008181MCP
get_claim(id=1008181)