Unverified50% confidenceFactExact time
ML4 is a model with 1 trillion total parameters and 52 billion active parameters, continuing Mistral's路线 on large-scale sparse architectures.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMistral Large 4 uses a MoE architecture with 1 trillion total parameters and 49 billion active parameters, is natively multimodal, and has a context window of 512,000 tokens.71% similarUnverifiedCommon on-device model deployment solutions include frameworks such as Core ML and llama.cpp, which can compress model weights to anywhere from a few hundred MB to several GB.68% similarUnverifiedAll four models have a context length of 1 million tokens.67% similarUnverifiedThe rough estimation formula for the memory (GB) required by a 4-bit quantized model is: parameter count (B) × 0.5 + 2.67% similarUnverified自托管 Mistral Large 4 按每参数 8 位计算，1.05 万亿参数需要约 1TB 内存，还不含 KV 缓存和运行时开销。67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1009648API
curl https://kongchang.com/api/v1/knowledge/claims/1009648MCP
get_claim(id=1009648)