Is Qwen3.8-9B-Q8_0 good for coding?

In our benchmarks, Qwen3.8-9B-Q8_0 scores 62.1/100 for coding. It runs at about 90.6 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-MLX-oQ4e is faster (~107.9 tok/s) and still scores well for coding (81.5/100).

Is Qwen3.8-9B-Q8_0 good for agentic (tool-using) tasks?

In our benchmarks, Qwen3.8-9B-Q8_0 scores 76.2/100 for agentic workflows. It runs at about 90.6 tok/s, so if you want more speed, qwen3.8-flash-next-iq3_s is faster (~116.8 tok/s) and still scores well for agentic workflows (88.4/100).

How fast is Qwen3.8-9B-Q8_0 for local inference?

Across 6 community benchmark runs, Qwen3.8-9B-Q8_0 reaches up to 127.6 tok/s and averages 90.6 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.

How much memory does Qwen3.8-9B-Q8_0 need?

The leanest observed configuration used about 10.2 GB of memory (quantizations tested: Q8_0).

Which tools have been used to run Qwen3.8-9B-Q8_0?

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

Model benchmarks

Qwen3.8-9B-Q8_0 local LLM performance

As of September 2026, Qwen3.8-9B-Q8_0 runs at up to 127.6 tok/s for local inference (best of 6 community benchmark runs across 2 GPUs).

Ollamallama.cppQ8_0

ShareReddit X

Model size

Peak speed

127.6 tok/s

Average speed

90.6 tok/s

Avg PP

1560.2 tok/s

Min memory

10.2 GB

Max context

131,072 tokens

Avg output / run

13,245 tokens

Avg runtime / run

2m 40s

Avg quality

72.6

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 6 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	67.9	72.6	77.4
Agent Workflow	60.8	76.2	89.4
Code Generation	55.9	62.1	67.8
Role Play & Narrative	66.1	76.8	85.3
Research & Analysis	74.2	75.5	77.6

Performance by hardware and tool

Every hardware/tool/quantization combination Qwen3.8-9B-Q8_0 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
AMD Radeon RX 7900 XTX	llama.cpp	Q8_0	127.6 tok/s	125.6 tok/s	n/a	65,536 tokens	68.3	2
AMD Radeon RX 7900 XTX	Ollama	Q8_0	76.5 tok/s	75.0 tok/s	10.2 GB	65,536 tokens	72.8	2
NVIDIA GeForce RTX 5060	llama.cpp	Q8_0	73.5 tok/s	71.2 tok/s	n/a	131,072 tokens	76.8	2

Benchmark runs

All 6 Qwen3.8-9B-Q8_0 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.

Open run page →

Frequently asked questions

Is Qwen3.8-9B-Q8_0 good for coding?: In our benchmarks, Qwen3.8-9B-Q8_0 scores 62.1/100 for coding. It runs at about 90.6 tok/s, so if you want more speed, Tiel-Coder-35B-A3B-MLX-oQ4e is faster (~107.9 tok/s) and still scores well for coding (81.5/100).
Is Qwen3.8-9B-Q8_0 good for agentic (tool-using) tasks?: In our benchmarks, Qwen3.8-9B-Q8_0 scores 76.2/100 for agentic workflows. It runs at about 90.6 tok/s, so if you want more speed, qwen3.8-flash-next-iq3_s is faster (~116.8 tok/s) and still scores well for agentic workflows (88.4/100).
How fast is Qwen3.8-9B-Q8_0 for local inference?: Across 6 community benchmark runs, Qwen3.8-9B-Q8_0 reaches up to 127.6 tok/s and averages 90.6 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
How much memory does Qwen3.8-9B-Q8_0 need?: The leanest observed configuration used about 10.2 GB of memory (quantizations tested: Q8_0).
Which tools have been used to run Qwen3.8-9B-Q8_0?: Benchmarks were submitted using Ollama, llama.cpp. Results are community-contributed and updated as new runs arrive.