← Back to community benchmarks
Meta-Llama-3.1-8B-Instruct
Performance
1k
tokens
233.5
PP tok/s
22.8
TG tok/s
4386
TTFT (ms)
4.9
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 233.5 | 22.8 | 4.9 GB | current |
| 4k | 215.8 | 20.5 | 5.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 22.8 | 1.00× |
| 2× | 43.8 | 1.92× |
| 4× | 51.3 | 2.25× |