Unverified50% confidenceBenchmarkExact time
The author mdnazib963 trained a 1.9 MB model from scratch that reaches an inference speed of 3405 tok/s on a single core of a Ryzen 5 3400G.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedA developer increased the inference speed of the LingBot-World 2.0 1.3B model on a single RTX 5090 from 6 FPS to 16 FPS and open-sourced the code.67% similarVerifiedIn benchmark testing, Qwen3 27B achieved a generation speed of approximately 40-something tokens per second on 4 modified 2080 Ti cards without NVLink enabled.67% similarUnverified开发者公布在RTX 3050 Laptop（4GB）上微调Llama-3.1-8B的实测结果为训练速度119.6 tok/s，峰值显存3.32 GB67% similarUnverifiedRunning the simulator alone on an A100 (65,536 environments) can achieve a throughput of 3.27 million env-steps/s.66% 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.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1006035API
curl https://kongchang.com/api/v1/knowledge/claims/1006035MCP
get_claim(id=1006035)