Unverified50% confidenceBenchmarkExact time
An RTX 4080S (Q4 quantization) used as a comparison can reach over 1,000 t/s during the preview phase.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/5/2026
First Seen
Valid until: 1/3/2027
Sources
Related Entities
Related Claims
UnverifiedThe typical generation speed of an NVIDIA RTX 4090 running a similarly scaled Q4 quantized model is in the range of 40–60 TPS74% similarUnverifiedGPT-OSS 20B在RTX 3090上可达约40 Token/s的生成速度，在RTX 5090上能达到200 Token/s73% similarUnverifiedA ~25 GB Q3 model running entirely within the RTX 4090's VRAM (bandwidth ~1 TB/s) theoretically achieves ~40 tokens per second, whereas running it in DDR5 memory (~80 GB/s) drops throughput to 1–3 tokens per second.71% similarUnverifiedRTX 4090相比RTX 3060在推理场景下的实际吞吐量提升约3-4倍69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/978759API
curl https://kongchang.com/api/v1/knowledge/claims/978759MCP
get_claim(id=978759)