Unverified50% confidenceBenchmarkExact time
According to the FreeToken technical paper, Qwen3-235B-A3B can achieve 77–83 tokens/s on an RTX 5090 system, and DeepSeek-related models can achieve 22–25 tokens/s on agent trajectories.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
UnverifiedWhen VRAM is sufficient and the model is fully resident on the GPU, an RTX 4090 running a Q4-quantized 27B model can typically achieve 15–30 tokens/s; once CPU offloading is triggered, the speed may drop sharply to 1–2 tokens/s or even lower.76% similarUnverifiedThe model can achieve approximately 206 tokens/second on a single RTX 5090 using the NVFP4 scheme.71% similarUnverifiedQwen3 27B features 64 layers, a hidden dimension of 5120, and a native context length of 262,144 tokens.70% similarUnverifiedQwen3 在训练时支持长达 128K tokens 的上下文68% similarUnverifiedThe context length of mainstream open-source multimodal models is typically between 4K and 32K tokens: LLaVA-1.5 is 4,096 tokens, Qwen-VL is 8,192 tokens, and InternVL2 extends to 32K.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/898238API
curl https://kongchang.com/api/v1/knowledge/claims/898238MCP
get_claim(id=898238)