← Back to community benchmarks
supergemma4-26b-abliterated-multimodal
Performance
4k
tokens
618.8
PP tok/s
57.2
TG tok/s
6619
TTFT (ms)
14.9
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 587.3 | 57.0 | 14.3 GB | view |
| 4k | 618.8 | 57.2 | 14.9 GB | current |
| 8k | 608.4 | 48.2 | 15.1 GB | view |
| 16k | 578.8 | 44.9 | 15.6 GB | view |
| 32k | 503.4 | 35.3 | 16.5 GB | view |
| 64k | 423.3 | 5.6 | 18.5 GB | view |
| 128k | 311.6 | 25.4 | 22.7 GB | view |
| 195k | 240.3 | 4.1 | 27.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 57.0 | 1.00× |
| 2× | 76.7 | 1.35× |
| 4× | 90.0 | 1.58× |
| 8× | 124.8 | 2.19× |