← Back to community benchmarks
GLM-4.7-8bit-gs32
Performance
32k
tokens
110.7
PP tok/s
6.6
TG tok/s
295991
TTFT (ms)
382.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.6
macOS
macOS 15.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 188.8 | 13.4 | 370.9 GB | view |
| 4k | 212.6 | 12.2 | 372.4 GB | view |
| 8k | 189.4 | 10.9 | 373.8 GB | view |
| 16k | 157.1 | 8.9 | 376.7 GB | view |
| 32k | 110.7 | 6.6 | 382.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 13.4 | 1.00× |
| 2× | 18.2 | 1.36× |