Unverified50% confidenceBenchmarkExact time
When running Nemotron Super (approximately 120 billion parameters), the unoptimized speed is about 16.6 tokens per second; after enabling MTP, Flash Attention, and KV cache, it improves to 27.2 tokens per second (at a context length of around 17,000), a speedup of approximately 64%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/25/2026
First Seen
Valid until: 12/24/2026
Sources
Related Entities
Related Claims
Unverified根据用户反馈，使用 Quantprobe 的深度感知优化配方，相比默认设置生成速度可以几乎翻倍，在 GTX 1060 上从约 10 tokens/s 跃升至超过 22 tokens/s65% similarVerifiedOn 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×.65% similarUnverifiedThe UP host tested MiniCPM-2B on a computer with 4G VRAM and achieved a generation speed of 57 tokens per second, with a first-token latency of only 0.22 seconds.64% similarUnverifiedThe theoretical peak of a 56k dial-up modem is approximately 7KB per second.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/950956API
curl https://kongchang.com/api/v1/knowledge/claims/950956MCP
get_claim(id=950956)