← Back to community benchmarks
Qwen3-VL-Embedding-2B
Performance
4k
tokens
744.7
PP tok/s
59.5
TG tok/s
5501
TTFT (ms)
2.7
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.2.6
macOS
macOS 26.3.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 818.5 | 63.9 | 2.5 GB | view |
| 4k | 744.7 | 59.5 | 2.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 63.9 | 1.00× |
| 2× | 92.1 | 1.44× |
| 4× | 129.5 | 2.03× |