Unverified50% confidenceFactExact time
SGLang admitted that its earlier comparison placing GB300 measured results side by side with DeepInfer's B300 data was unfair, and after retesting said performance had matched the DeepInfer benchmark, with decoding throughput improving 7.4% on a 4-GB300 configuration.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/25/2026
First Seen
Valid until: 1/23/2027
Sources
Related Entities
Related Claims
UnverifiedIn the same test, the bf16 baseline exceeded the self-developed W4 kernel (25.5 tok/s) and bitsandbytes NF4 (25.7 tok/s) at 31.3 tok/s, reaching 29.8% of the theoretical roofline.66% similarUnverifiedDeepSeek-7B-Chat's precision drops sharply to 35.3%, with its principal components dominated by structural variance rather than style variance.65% similarUnverified在覆盖九大维度共403道题的测试中，Muse-Glimmer-30B综合均分为76.04，比Corpus 3.6 27B的75.57高出0.47分65% similarUnverifiedOn the B300 benchmark, Kimi K3's new optimization solution on vLLM achieved a 2.2 to 2.8× throughput improvement compared to version v0.27.1.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1095773API
curl https://kongchang.com/api/v1/knowledge/claims/1095773MCP
get_claim(id=1095773)