Is qwen3.8:latest good for coding?

In our benchmarks, qwen3.8:latest scores 76.3/100 for coding. It runs at about 37.9 tok/s, so if you want more speed, K2-Horizon-MoVA-36B-A4B-MLX-4bit is faster (~48.5 tok/s) and still scores well for coding (82.5/100).

Is qwen3.8:latest good for agentic (tool-using) tasks?

In our benchmarks, qwen3.8:latest scores 81.3/100 for agentic workflows. It runs at about 37.9 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).

How fast is qwen3.8:latest for local inference?

Across 8 community benchmark runs, qwen3.8:latest reaches up to 42.0 tok/s and averages 37.9 tok/s, with the fastest results on NVIDIA GeForce RTX 5060 Ti.

How much memory does qwen3.8:latest need?

The leanest observed configuration used about 16.3 GB of memory (quantizations tested: Q4_K_M).

Which tools have been used to run qwen3.8:latest?

Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.

Model benchmarks

qwen3.8:latest local LLM performance

As of September 2026, qwen3.8:latest runs at up to 42.0 tok/s for local inference (best of 8 community benchmark runs across 2 GPUs).

OllamaQ4_K_M

ShareReddit X

Model size

27.3B

Peak speed

42.0 tok/s

Average speed

37.9 tok/s

Avg PP

436.4 tok/s

Min memory

16.3 GB

Max context

163,840 tokens

Avg output / run

16,428 tokens

Avg runtime / run

8m 21s

Avg quality

82.5

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 8 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	80.5	82.5	84.7
Agent Workflow	71.8	81.3	87.8
Code Generation	67.3	76.3	80.8
Role Play & Narrative	73.9	85.6	92.8
Research & Analysis	83.1	86.7	89.3

Performance by hardware and tool

Every hardware/tool/quantization combination qwen3.8:latest 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 5060 Ti	Ollama	Q4_K_M	42.0 tok/s	42.0 tok/s	17.1 GB	8,192 tokens	81.6	1
AMD Radeon RX 7900 XTX	Ollama	Q4_K_M	41.9 tok/s	37.4 tok/s	16.3 GB	163,840 tokens	82.6	7

Benchmark runs

All 8 qwen3.8:latest runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Open run page →

Frequently asked questions

Is qwen3.8:latest good for coding?: In our benchmarks, qwen3.8:latest scores 76.3/100 for coding. It runs at about 37.9 tok/s, so if you want more speed, K2-Horizon-MoVA-36B-A4B-MLX-4bit is faster (~48.5 tok/s) and still scores well for coding (82.5/100).
Is qwen3.8:latest good for agentic (tool-using) tasks?: In our benchmarks, qwen3.8:latest scores 81.3/100 for agentic workflows. It runs at about 37.9 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).
How fast is qwen3.8:latest for local inference?: Across 8 community benchmark runs, qwen3.8:latest reaches up to 42.0 tok/s and averages 37.9 tok/s, with the fastest results on NVIDIA GeForce RTX 5060 Ti.
How much memory does qwen3.8:latest need?: The leanest observed configuration used about 16.3 GB of memory (quantizations tested: Q4_K_M).
Which tools have been used to run qwen3.8:latest?: Benchmarks were submitted using Ollama. Results are community-contributed and updated as new runs arrive.