Expired60% confidenceFactTime unknown
For the RTX 4090 (24GB VRAM), WhichLLM may recommend the Q5 quantized version of Qwen3.6 27B, with an estimated inference speed of approximately 27 tokens per second.
2
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/1/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedQwen2.5-Coder-32B running with Q4 quantization on an RTX 4090 can achieve an inference speed of 40–60 tokens per second.83% similarUnverifiedA single RTX 4090 with 24 GB of VRAM can run the 4-bit quantized version of Qwen3-32B smoothly.82% similarVerifiedQLoRA introduces 4-bit quantization technology, making it possible to fine-tune models with 65 billion parameters on a single RTX 3090/4090 GPU with 24 GB of VRAM.79% similarUnverifiedOn the RTX 5090, the Qwen3.6 27B model's prefill speed increases from 174 to 253 tokens/s, with a VRAM usage of approximately 18 GB.78% similarUnverifiedWhen 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.77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/38146API
curl https://kongchang.com/api/v1/knowledge/claims/38146MCP
get_claim(id=38146)