← Back to community benchmarks
GLM-54.8bit
Performance
16k
tokens
119.3
PP tok/s
12.2
TG tok/s
137303
TTFT (ms)
430.9
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
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 | 178.6 | 16.2 | 420.0 GB | view |
| 4k | 180.5 | 13.5 | 422.3 GB | view |
| 8k | 155.8 | 13.1 | 424.9 GB | view |
| 16k | 119.3 | 12.2 | 430.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.2 | 1.00× |
| 2× | 22.9 | 1.41× |
| 4× | 30.0 | 1.85× |
| 8× | 41.1 | 2.54× |