qwen3-coder-bench-32k on M4 Max (40-core GPU) + 128GB unified
MeasuredOn 2026-06-27, qwen3-coder-bench-32k Q4_K_M ran at 113.3 tok/s (chat-short) on an M4 Max (40-core GPU) with ollama. Run r_roktphpc--8.
M4 Max (40-core GPU) + 128GB unified
suite suite-v1
cli 0.0.3
signedya0OvDcfH4…
Embed badgesubmitted Jun 27, 2026Workload results
| Workload | Backend | Model | decode tok/s | prefill tok/s | TTFT | p50 | p95 |
|---|---|---|---|---|---|---|---|
| chat-short | ollama@0.30.11 | qwen3-coder-bench-32kQ4_K_M | 113.3tok/s | 84.12tok/s | 1,308ms | 8.9ms | 9.2ms |
| chat-long | ollama@0.30.11 | qwen3-coder-bench-32kQ4_K_M | 97.40tok/s | 1,342.8tok/s | 2,344ms | 10.3ms | 10.8ms |
| concurrent-decode | ollama@0.30.11 | qwen3-coder-bench-32kQ4_K_M | 109.7tok/s | — | — | 9.1ms | 9.5ms |
| agent-trace | ollama@0.30.11 | qwen3-coder-bench-32kQ4_K_M | 103.2tok/s | 3,371.6tok/s | 477ms | 9.5ms | 10.5ms |
Reproduce on your machine
Start with this model and workload selection. This command does not capture the original artifact, backend version or server settings:
Match quantization, context capacity, caching and runtime settings for a useful comparison. Duration varies by workload and hardware; the published client may differ from this run's version. How it's measured.
Share this run
Copy the short link, a Reddit-ready summary, or a README badge. Every number is the measured result on this page, and each badge links back to it.
https://llm-speed.com/r/r_roktphpc--8Related benchmarks
- More qwen3-coder-bench-32k benchmarks — every backend and rig that has run this model.
- More M4 Max (40-core GPU) LLM benchmarks — every model measured on this hardware.
Provenance
- Run ID
- r_roktphpc--8
- Fingerprint hash
- 275ecb2b79296aab
- Public key
- ya0OvDcfH4La0mEhEhM8iESwvi9/MZz9uibPMNfpovE=
- Received
- 2026-06-27 16:14:03