Unverified50% confidenceBenchmarkExact time
On an 8x H200/B200 node, Mistral Large 3 achieves a sustained generation speed of 142 tokens/second.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/12/2026
First Seen
Valid until: 10/26/2026
Sources
Related Entities
Related Claims
VerifiedOn the RTX 5090, Muse Glimmer's generation speed increases from 75 tokens per second to 233 tokens per second, a speedup of approximately 3.1×.72% similarUnverifiedThe M5 Max's token generation speed is approximately in the range of 10–30 tokens/s (depending on model size), while the NVIDIA H100's HBM3 memory bandwidth reaches up to 3.35 TB/s, enabling inference speeds tens of times faster than local devices.70% similarUnverified在配备96GB显存的M2 Max上以1-bit精度运行时，简单任务生成速度稳定在约28-30 tokens/秒69% similarUnverifiedWith an approximately 1.6M ultra-long context and 128 concurrent requests, throughput can reach approximately 3,700 tokens/s (aggregated).67% similarUnverifiedOn an RTX 5090, Muse Glimmer runs at 75 tokens/second without speculative decoding, and reaches 233 tokens/second when paired with a speculative decoding model67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1008186API
curl https://kongchang.com/api/v1/knowledge/claims/1008186MCP
get_claim(id=1008186)