Model benchmarks
unsloth/Qwen3.8-Flash-Next-GGUF local LLM performance
As of September 2026, unsloth/Qwen3.8-Flash-Next-GGUF runs at up to 43.4 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
Model size
Unknown
Peak speed
43.4 tok/s
Average speed
36.2 tok/s
Avg PP
245.1 tok/s
Min memory
n/a
Max context
65,536 tokens
Avg output / run
53,875 tokens
Avg runtime / run
26m 40s
Avg quality
76.6
Benchmark runs
4
GPUs tested
2
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 | 50.5 | 76.6 | 87.9 |
| Agent Workflow | 79.8 | 88.4 | 93.1 |
| Code Generation | 17.7 | 58.9 | 79.3 |
| Role Play & Narrative | 90.6 | 92.3 | 94.0 |
| Research & Analysis | 13.1 | 66.7 | 90.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/Qwen3.8-Flash-Next-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 5090 | vLLM | — | 43.4 tok/s | 39.3 tok/s | n/a | 65,536 tokens | 87.4 | 3 |
| AMD Radeon RX 7900 XTX | Unsloth Studio | UD-IQ3_XXS | 27.1 tok/s | 27.1 tok/s | n/a | 16,384 tokens | 44.1 | 1 |
Benchmark runs
All 4 unsloth/Qwen3.8-Flash-Next-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/Qwen3.8-Flash-Next-GGUF good for coding?
- In our benchmarks, unsloth/Qwen3.8-Flash-Next-GGUF scores 58.9/100 for coding. It runs at about 36.2 tok/s, so if you want more speed, local-qwen is faster (~60.2 tok/s) and still scores well for coding (84.8/100).
- Is unsloth/Qwen3.8-Flash-Next-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/Qwen3.8-Flash-Next-GGUF scores 88.4/100 for agentic workflows. It runs at about 36.2 tok/s, so if you want more speed, granite-4.2-8b-Q4_K_L is faster (~71.6 tok/s) and still scores well for agentic workflows (92.9/100).
- How fast is unsloth/Qwen3.8-Flash-Next-GGUF for local inference?
- Across 4 community benchmark runs, unsloth/Qwen3.8-Flash-Next-GGUF reaches up to 43.4 tok/s and averages 36.2 tok/s, with the fastest results on NVIDIA GeForce RTX 5090.
- Which tools have been used to run unsloth/Qwen3.8-Flash-Next-GGUF?
- Benchmarks were submitted using Unsloth Studio, vLLM. Results are community-contributed and updated as new runs arrive.