Model benchmarks
unsloth/Qwen3.5-4B-MTP-GGUF local LLM performance
As of September 2026, unsloth/Qwen3.5-4B-MTP-GGUF runs at up to 146.2 tok/s for local inference (best of 4 community benchmark runs across 1 GPU).
Model size
4B
Peak speed
146.2 tok/s
Average speed
103.9 tok/s
Avg PP
1578.0 tok/s
Min memory
11.2 GB
Max context
262.144 tokens
Avg output / run
9.098 tokens
Avg runtime / run
2m 9s
Avg quality
59.9
Benchmark runs
4
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 4 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 | 54.4 | 59.9 | 68.3 |
| Agent Workflow | 26.4 | 52.0 | 83.0 |
| Code Generation | 50.0 | 53.9 | 57.7 |
| Role Play & Narrative | 61.6 | 67.7 | 74.7 |
| Research & Analysis | 60.2 | 66.0 | 72.4 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/Qwen3.5-4B-MTP-GGUF has been benchmarked on, ranked by peak token generation speed. Last updated September 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 4070 | Unsloth Studio | UD-Q4_K_XL | 146.2 tok/s | 133.6 tok/s | 17.1 GB | 65.536 tokens | 63.8 | 2 |
| NVIDIA GeForce RTX 4070 | Unsloth Studio | — | 123.0 tok/s | 123.0 tok/s | 11.2 GB | 65.536 tokens | 53.8 | 1 |
| NVIDIA GeForce RTX 4070 | Unsloth Studio | UD-Q6_K_XL | 25.5 tok/s | 25.5 tok/s | 18.3 GB | 262.144 tokens | 58.1 | 1 |
Benchmark runs
All 4 unsloth/Qwen3.5-4B-MTP-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/Qwen3.5-4B-MTP-GGUF good for coding?
- In our benchmarks, unsloth/Qwen3.5-4B-MTP-GGUF scores 53.9/100 for coding. It runs at about 103.9 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 unsloth/Qwen3.5-4B-MTP-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/Qwen3.5-4B-MTP-GGUF scores 52.0/100 for agentic workflows. It runs at about 103.9 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-Q4_K_S is faster (~162.4 tok/s) and still scores well for agentic workflows (87.8/100).
- How fast is unsloth/Qwen3.5-4B-MTP-GGUF for local inference?
- Across 4 community benchmark runs, unsloth/Qwen3.5-4B-MTP-GGUF reaches up to 146.2 tok/s and averages 103.9 tok/s, with the fastest results on NVIDIA GeForce RTX 4070.
- How much memory does unsloth/Qwen3.5-4B-MTP-GGUF need?
- The leanest observed configuration used about 11.2 GB of memory (quantizations tested: UD-Q4_K_XL, UD-Q6_K_XL).
- Which tools have been used to run unsloth/Qwen3.5-4B-MTP-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.