← Back to community benchmarks
GLM-4.7-Flash-abliterated-mxfp8
Performance
16k
tokens
780.6
PP tok/s
36.9
TG tok/s
20990
TTFT (ms)
34.0
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,207 | 53.3 | 29.4 GB | view |
| 4k | 1,298 | 50.3 | 30.4 GB | view |
| 8k | 1,081 | 44.6 | 31.6 GB | view |
| 16k | 780.6 | 36.9 | 34.0 GB | current |
| 32k | 492.7 | 28.5 | 39.0 GB | view |
| 64k | 273.0 | 18.3 | 49.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 53.3 | 1.00× |
| 2× | 84.0 | 1.58× |
| 4× | 119.5 | 2.24× |
| 8× | 152.8 | 2.87× |