Unverified50% confidenceFactExact time
MNN-Pocket-0.3B has a deployment size of 173MB under INT4 quantization.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/10/2026
First Seen
Valid until: 2/8/2027
Sources
Related Entities
Related Claims
UnverifiedM3原始权重存储下限:BF16约854GB,8-bit约428GB,4-bit约214GB70% similarVerifiedA 70B-parameter FP16 model requires approximately 140 GB of video memory, approximately 70 GB after INT8 quantization, and approximately 35–40 GB after INT4 quantization.68% similarVerifiedMiniMax H3 still requires close to 24 GB of VRAM under NF4 quantization.68% similarUnverifiedIn an experimental configuration, Ollama identifies the 840M as having approximately 16 GB of VRAM (16193 MiB), with about 14.6 GiB available for the Vulkan device at load time.67% similarVerifiedA 70B model can be compressed to approximately 35GB at INT4 precision.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1154759API
curl https://kongchang.com/api/v1/knowledge/claims/1154759MCP
get_claim(id=1154759)