Is unsloth/Qwen3.5-4B-MTP-GGUF good for agentic (tool-using) tasks?

In our benchmarks, unsloth/Qwen3.5-4B-MTP-GGUF scores 59.4/100 for agentic workflows. It runs at about 108.4 tok/s, so if you want more speed, qwen3.8-flash-next-q2_0 is faster (~127.9 tok/s) and still scores well for agentic workflows (85.6/100).

How fast is unsloth/Qwen3.5-4B-MTP-GGUF for local inference?

Across 21 community benchmark runs, unsloth/Qwen3.5-4B-MTP-GGUF reaches up to 149.8 tok/s and averages 108.4 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 7.2 GB of memory (quantizations tested: Q8_0, 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.

Model benchmarks

unsloth/Qwen3.5-4B-MTP-GGUF local LLM performance

As of October 2026, unsloth/Qwen3.5-4B-MTP-GGUF runs at up to 149.8 tok/s for local inference (best of 21 community benchmark runs across 2 GPUs).

Unsloth StudioQ8_0UD-Q4_K_XLUD-Q6_K_XL

ShareReddit X

Model size

Peak speed

149.8 tok/s

Average speed

108.4 tok/s

Avg PP

1477.7 tok/s

Min memory

7.2 GB

Max context

262,144 tokens

Avg output / run

16,019 tokens

Avg runtime / run

3m 24s

Avg quality

64.3

Benchmark runs

GPUs tested

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 21 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	56.9	64.3	70.5
Agent Workflow	42.1	59.4	74.0
Code Generation	38.9	54.4	66.2
Role Play & Narrative	61.5	74.0	84.0
Research & Analysis	59.7	69.5	77.3

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 October 2026.

Hardware	Tool	Quant	Best tok/s	Avg tok/s	Memory	Context	Quality	Runs
NVIDIA GeForce RTX 4070	Unsloth Studio	UD-Q4_K_XL	149.8 tok/s	138.8 tok/s	7.2 GB	65,536 tokens	65.6	6
NVIDIA GeForce RTX 4070	Unsloth Studio	—	123.0 tok/s	123.0 tok/s	n/a	65,536 tokens	53.8	1
NVIDIA GeForce RTX 4070	Unsloth Studio	UD-Q6_K_XL	122.8 tok/s	100.3 tok/s	9.1 GB	262,144 tokens	64.8	13
NVIDIA GeForce RTX 3070 Ti	Unsloth Studio	Q8_0	15.9 tok/s	15.9 tok/s	n/a	8,192 tokens	61.2	1

Benchmark runs

All 21 unsloth/Qwen3.5-4B-MTP-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Open run page →

Frequently asked questions

Is unsloth/Qwen3.5-4B-MTP-GGUF good for agentic (tool-using) tasks?: In our benchmarks, unsloth/Qwen3.5-4B-MTP-GGUF scores 59.4/100 for agentic workflows. It runs at about 108.4 tok/s, so if you want more speed, qwen3.8-flash-next-q2_0 is faster (~127.9 tok/s) and still scores well for agentic workflows (85.6/100).
How fast is unsloth/Qwen3.5-4B-MTP-GGUF for local inference?: Across 21 community benchmark runs, unsloth/Qwen3.5-4B-MTP-GGUF reaches up to 149.8 tok/s and averages 108.4 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 7.2 GB of memory (quantizations tested: Q8_0, 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.