← Back to community benchmarks
GLM-OCR
Performance
8k
tokens
3,186
PP tok/s
119.4
TG tok/s
2571
TTFT (ms)
3.2
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
64 GB
GPU Cores
24
Software
oMLX
v0.2.10
macOS
macOS 26.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 4,355 | 176.3 | 2.9 GB | view |
| 4k | 3,936 | 145.8 | 3.0 GB | view |
| 8k | 3,186 | 119.4 | 3.2 GB | current |
| 32k | 1,504 | 62.6 | 4.6 GB | view |
| 64k | 875.7 | 40.1 | 6.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 176.3 | 1.00× |
| 2× | 101.2 | 0.57× |
| 4× | 188.6 | 1.07× |
| 8× | 313.3 | 1.78× |