Unverified50% confidenceBenchmarkExact time
Qwen3 27B's local inference speed is approximately 10 tokens per second.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
VerifiedQwen2.5-Coder-32B running with Q4 quantization on an RTX 4090 can achieve an inference speed of 40–60 tokens per second.77% similarUnverifiedIn benchmark testing, Qwen3 27B achieved a generation speed of approximately 40-something tokens per second on 4 modified 2080 Ti cards without NVLink enabled.77% similarUnverifiedQwythos-9B achieves a token output speed of approximately 75 tokens per second under Q8 quantization.73% similarVerifiedIn an RTX 5070 (12GB VRAM) configuration, the local Qwen3 9B model's output speed reached approximately 70–76 tokens/second.71% similarUnverifiedOn a dual AGX Orin system, Qwen 3.8 Flash Next achieves an inference speed of 38 tokens/second.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/907338API
curl https://kongchang.com/api/v1/knowledge/claims/907338MCP
get_claim(id=907338)