Unverified50% confidenceBenchmarkExact time
In actual benchmark threads for llama.cpp, the 2021 M1 Max achieved a decode write speed of 61 tokens/s, higher than the 2024 M4 Pro's 51 tokens/s.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe Neural Accelerator requires supporting software to function; running M5 Max with a 2023 old version of llama.cpp yields a read speed of only approximately 990 tokens/s.76% similarUnverifiedWhen running Qwen3-Next-Flash (Autoround W4A16 quantization), the prefill speed was approximately 1.3k tokens/s, and the generation speed reached 70 tokens/s (code) / 60 tokens/s (prose), with support for 128k+ long context.74% similarUnverifiedUnder this configuration, the 35B model achieves a measured single-user peak speed of approximately 80 tokens/second, with concurrent throughput exceeding 500 tokens/second.74% similarUnverifiedBased on the MLX framework, top-configured Mac models can achieve a generation speed of approximately 20 tokens/second when running the Llama-3-70B quantized model.72% similarUnverifiedIn a text generation test using an 8-billion-parameter model, the M5 Ultra achieved 188 tokens per second versus 129 tokens per second for the M3 Ultra, representing only 1.45x.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/977841API
curl https://kongchang.com/api/v1/knowledge/claims/977841MCP
get_claim(id=977841)