Qwen2.5-Coder-32B-Instruct vs Codestral-22B-v0.1
Side-by-side decode tok/s, prefill tok/s, and TTFT for Qwen2.5-Coder-32B-Instruct and Codestral-22B-v0.1, sourced from community-submitted runs of the llm-speed suite. Every number on this page links back to the run it came from.
Verdict
On the M3 Ultra (60-core GPU), Codestral-22B-v0.1 decodes at 47 tok/s versus 34 tok/s for Qwen2.5-Coder-32B-Instruct, 1.4× faster. That is the only hardware measured on both so far; the row below links to both source runs. Submit another to widen the comparison.
Hardware with data on both models
| Hardware | Qwen2.5-Coder-32B-Instruct decode | Codestral-22B-v0.1 decode | Δ | Source runs |
|---|---|---|---|---|
| M3 Ultra (60-core GPU) | 34.48tok/s | 47.49tok/s | -13.0 | r_721b4bls_oq · r_79dvtag5fd_ |
See also: Qwen2.5-Coder-32B-Instruct benchmarks · Codestral-22B-v0.1 benchmarks · All hardware · All models · Methodology