Model benchmarks
Qwen3.5:4B/MTP@262K local LLM performance
As of August 2026, Qwen3.5:4B/MTP@262K runs at up to 362.6 tok/s for local inference (best of 2 community benchmark runs across 0 GPUs).
LM Studio
Model size
4B
Peak speed
362.6 tok/s
Average speed
280.9 tok/s
Min memory
1000 MB
Max context
14,047 tokens
Best quality
76.5
Benchmark runs
2
GPUs tested
0
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.5:4B/MTP@262K has been benchmarked on, ranked by peak token generation speed. Last updated August 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| CPU only | LM Studio | — | 362.6 tok/s | 280.9 tok/s | 1000 MB | 14,047 tokens | 34.1 | 2 |
Frequently asked questions
- How fast is Qwen3.5:4B/MTP@262K for local inference?
- Across 2 community benchmark runs, Qwen3.5:4B/MTP@262K reaches up to 362.6 tok/s and averages 280.9 tok/s.
- How much memory does Qwen3.5:4B/MTP@262K need?
- The leanest observed configuration used about 1000 MB of memory.
- Which tools have been used to run Qwen3.5:4B/MTP@262K?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.