Compare
Loading community benchmarks…
Loading community benchmarks…
Put two chips, runtimes, or runtime versions side by side. Everything else is held fixed or called out, so a ratio is worth exactly what the matched runs say.
Ratios read side A relative to side B. Geometric mean over 2 like-for-like pairs.
Everything else matches: model, quantisation, backend, protocol, and workload are the same on both sides.
Each row is a configuration present on both sides. Values are per-cell medians; ratios read Apple M1 Max relative to Apple M3 Ultra.
| Configuration | Decode A | Decode B | Ratio | Prefill A | Prefill B | Ratio | Runs A / B |
|---|---|---|---|---|---|---|---|
Qwen3-4B BaseRTQ4 | 103.1 | 168.3 | 0.61× | 926 | 2,526 | 0.37× | 2 / 2 |
Qwen3-4B BaseRTQ8 | 71.3 | 119.7 | 0.60× | 919 | 2,492 | 0.37× | 2 / 2 |
Top 4 of 4 runs by decode throughput.
| Model / format | Chip / backend | Decode | Prefill | Contributor | Date | Report |
|---|---|---|---|---|---|---|
Qwen3-4B-Q4 BaseRT 0.3.0Q4 | Apple M1 Max Metal | 104.3 TG128 @ 1 ctx | 926 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q4 BaseRT 0.2.6Q4 | Apple M1 Max Metal | 101.9 TG128 @ 1 ctx | 927 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q8 BaseRT 0.2.6Q8 | Apple M1 Max Metal | 71.3 TG128 @ 1 ctx | 920 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q8 BaseRT 0.3.0Q8 | Apple M1 Max Metal | 71.2 TG128 @ 1 ctx | 918 PP512 | basecompute | View Benchmark |
Top 4 of 4 runs by decode throughput.
| Model / format | Chip / backend | Decode | Prefill | Contributor | Date | Report |
|---|---|---|---|---|---|---|
Qwen3-4B-Q4 BaseRT 0.3.0Q4 | Apple M3 Ultra Metal | 170.9 TG128 @ 1 ctx | 2,591 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q4 BaseRT 0.2.6Q4 | Apple M3 Ultra Metal | 165.7 TG128 @ 1 ctx | 2,461 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q8 BaseRT 0.3.0Q8 | Apple M3 Ultra Metal | 122.5 TG128 @ 1 ctx | 2,542 PP512 | basecompute | View Benchmark | |
Qwen3-4B-Q8 BaseRT 0.2.6Q8 | Apple M3 Ultra Metal | 116.8 TG128 @ 1 ctx | 2,441 PP512 | basecompute | View Benchmark |