Model benchmarks
ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS local LLM performance
As of November 2026, ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS runs at up to 54.2 tok/s for local inference (best of 8 community benchmark runs across 1 GPU).
Model size
177B
Peak speed
54.2 tok/s
Average speed
48.1 tok/s
Avg PP
228.7 tok/s
Min memory
31.9 GB
Max context
262,144 tokens
Avg output / run
41,056 tokens
Avg runtime / run
15m 30s
Avg quality
84.9
Benchmark runs
8
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 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 | 80.9 | 84.9 | 88.7 |
| Agent Workflow | 82.2 | 87.8 | 92.2 |
| Code Generation | 66.6 | 74.4 | 82.2 |
| Role Play & Narrative | 80.5 | 90.1 | 95.5 |
| Research & Analysis | 82.7 | 87.2 | 91.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS has been benchmarked on, ranked by peak token generation speed. Last updated November 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| Apple M4 Pro | OpenAI-compatible | — | 54.2 tok/s | 48.1 tok/s | 31.9 GB | 262,144 tokens | 84.9 | 8 |
Benchmark runs
All 8 ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS good for coding?
- In our benchmarks, ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS scores 74.4/100 for coding. It runs at about 48.1 tok/s, so if you want more speed, turbo is faster (~124.1 tok/s) and still scores well for coding (85.3/100).
- Is ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS good for agentic (tool-using) tasks?
- In our benchmarks, ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS scores 87.8/100 for agentic workflows. It runs at about 48.1 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 ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS for local inference?
- Across 8 community benchmark runs, ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS reaches up to 54.2 tok/s and averages 48.1 tok/s, with the fastest results on Apple M4 Pro.
- How much memory does ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS need?
- The leanest observed configuration used about 31.9 GB of memory.
- Which tools have been used to run ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF:IQ3_XXS?
- Benchmarks were submitted using OpenAI-compatible. Results are community-contributed and updated as new runs arrive.