Model benchmarks
Qwen3.6-35B-A3B-UD-IQ2_M local LLM performance
As of November 2026, Qwen3.6-35B-A3B-UD-IQ2_M runs at up to 113.8 tok/s for local inference (best of 2 community benchmark runs across 0 GPUs).
Model size
35B
Peak speed
113.8 tok/s
Average speed
110.2 tok/s
Avg PP
1356.1 tok/s
Min memory
15.4 GB
Max context
98,304 tokens
Avg output / run
16,641 tokens
Avg runtime / run
2m 36s
Avg quality
78.6
Benchmark runs
2
GPUs tested
0
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 | 77.2 | 78.6 | 80.0 |
| Agent Workflow | 71.9 | 77.3 | 82.7 |
| Code Generation | 72.7 | 73.5 | 74.3 |
| Role Play & Narrative | 81.5 | 82.4 | 83.3 |
| Research & Analysis | 79.1 | 81.1 | 83.2 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.6-35B-A3B-UD-IQ2_M has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon Graphics | llama.cpp | — | 113.8 tok/s | 110.2 tok/s | 15.4 GB | 98,304 tokens | 78.6 | 2 |
Benchmark runs
All 2 Qwen3.6-35B-A3B-UD-IQ2_M runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.6-35B-A3B-UD-IQ2_M good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.6-35B-A3B-UD-IQ2_M scores 77.3/100 for agentic workflows. It runs at about 110.2 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-MLX-oQ4e-MTP is faster (~151.8 tok/s) and still scores well for agentic workflows (83.0/100).
- How fast is Qwen3.6-35B-A3B-UD-IQ2_M for local inference?
- Across 2 community benchmark runs, Qwen3.6-35B-A3B-UD-IQ2_M reaches up to 113.8 tok/s and averages 110.2 tok/s, with the fastest results on AMD Radeon Graphics.
- How much memory does Qwen3.6-35B-A3B-UD-IQ2_M need?
- The leanest observed configuration used about 15.4 GB of memory.
- Which tools have been used to run Qwen3.6-35B-A3B-UD-IQ2_M?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.