← Back to community benchmarks
Ornith-1.0-35B
Performance
8k
tokens
2,824
PP tok/s
70.8
TG tok/s
2901
TTFT (ms)
36.8
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.5.1
macOS
macOS 26.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,456 | 73.8 | 35.4 GB | view |
| 4k | 2,927 | 73.2 | 36.2 GB | view |
| 8k | 2,824 | 70.8 | 36.8 GB | current |
| 16k | 2,577 | 68.7 | 37.5 GB | view |
| 32k | 2,185 | 64.8 | 39.5 GB | view |
| 64k | 1,624 | 56.7 | 42.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 73.8 | 1.00× |
| 2× | 134.4 | 1.82× |
| 4× | 216.3 | 2.93× |
| 8× | 343.4 | 4.65× |