Unverified50% confidenceBenchmarkExact time
On an M4 Max MacBook Pro (36GB unified memory), the MLX-format Qwen 27B sustains a decoding speed of approximately 40–41 tok/s in thermal measurements.
1
Sources
50%
Confidence
Long-term
Relevance
10/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverified27B稠密模型在24GB M4 Pro Mac mini上使用4-bit量化可以运行，速度约为11 tok/s75% similarUnverifiedThe 4-bit quantized weights of Qwen3 27B are only approximately 16.1 GB, allowing it to run on a MacBook Pro/Mac Studio with 32 GB of memory or a consumer-grade GPU with approximately 17 GB of VRAM.75% similarUnverifiedOn an M4 24GB device, a 9B model (such as qwen3.5-32k:9b-mlx) runs fast with low memory usage, but is unable to effectively identify code issues.73% similarUnverifiedOn a 128GB MacBook Pro M3 Max, independent benchmarks show DS4 achieving a generation speed of 26.68 tokens per second.71% similarUnverifiedRunning a Qwen3 27B-level model on an AMD RX 7900XTX 24GB GPU, the output speed can reach up to 52–53 TPS, with an average stable speed of around 47 TPS.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/968803API
curl https://kongchang.com/api/v1/knowledge/claims/968803MCP
get_claim(id=968803)