Expired50% confidenceFactExact time
Hunyuan V3 has open-sourced a BF16 version and also provides an FP8 quantized version, with the FP8 version being approximately 300 GB.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/10/2026
First Seen
Valid until: 10/8/2026(expired)
Sources
Related Claims
UnverifiedThe Qwen 3.6 27B FP8 model was successfully deployed locally on a 4×3080Ti 16GB GPU.72% similarUnverifiedA developer locally deployed the Qwen 3.6 27B FP8 model using 4 modified 3080Ti 16G graphics cards for programming benchmarking.70% similarUnverifiedA 2.4-trillion-parameter model requires approximately 9.6 TB of VRAM in FP32, about 4.8 TB in BF16, about 2.4 TB in INT8, and about 1.2 TB in INT4.68% similarUnverifiedThe B200's FP4 peak compute can reach approximately 9 PFLOPS (sparse mode), representing up to an 8x improvement over FP16.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/465613API
curl https://kongchang.com/api/v1/knowledge/claims/465613MCP
get_claim(id=465613)