Model benchmarks
Qwen3.6:35B-4bit local LLM performance
As of July 2026, Qwen3.6:35B-4bit runs at up to 103.4 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
oMLX
Model size
35B
Peak speed
103.4 tok/s
Average speed
64.6 tok/s
Min memory
21.5 GB
Max context
65,536 tokens
Avg quality
80.9
Benchmark runs
2
GPUs tested
1
Quality by task
Average LLM-judged quality (0–100) with the run-to-run spread shown as a P5–P95 band, overall and for each benchmark task, across all 2 runs. The low and high columns show how much the judge’s score varies between runs, and need at least two runs to display.
| Task | P5 (low) | Avg | P95 (high) |
|---|---|---|---|
| Overall | 79.6 | 80.9 | 82.1 |
| Agent Workflow | 80.7 | 83.3 | 85.8 |
| Code Generation | 74.5 | 77.3 | 80.1 |
| Role Play & Narrative | 75.7 | 80.9 | 86.1 |
| Research & Analysis | 81.5 | 82.0 | 82.5 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.6:35B-4bit has been benchmarked on, ranked by peak token generation speed. Last updated July 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M5 Max | oMLX | — | 103.4 tok/s | 64.6 tok/s | 21.5 GB | 65,536 tokens | 80.9 | 2 |
Frequently asked questions
- Is Qwen3.6:35B-4bit good for coding?
- In our benchmarks, Qwen3.6:35B-4bit scores 77.3/100 for coding. It runs at about 64.6 tok/s, so if you want more speed, Qwen3.6-35B-A3B-Q4_K_XL is faster (~156.9 tok/s) and still scores well for coding (82.2/100).
- Is Qwen3.6:35B-4bit good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.6:35B-4bit scores 83.3/100 for agentic workflows. It runs at about 64.6 tok/s, so if you want more speed, unsloth/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF:UD-Q4_K_XL is faster (~169.2 tok/s) and still scores well for agentic workflows (83.0/100).
- How fast is Qwen3.6:35B-4bit for local inference?
- Across 2 community benchmark runs, Qwen3.6:35B-4bit reaches up to 103.4 tok/s and averages 64.6 tok/s, with the fastest results on Apple M5 Max.
- How much memory does Qwen3.6:35B-4bit need?
- The leanest observed configuration used about 21.5 GB of memory.
- Which tools have been used to run Qwen3.6:35B-4bit?
- Benchmarks were submitted using oMLX. Results are community-contributed and updated as new runs arrive.