Model benchmarks
qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram local LLM performance
As of November 2026, qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram runs at up to 72.5 tok/s for local inference (best of 5 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
72.5 tok/s
Average speed
71.2 tok/s
Avg PP
278.6 tok/s
Min memory
80.7 GB
Max context
262,144 tokens
Avg output / run
62,077 tokens
Avg runtime / run
14m 53s
Avg quality
86.2
Benchmark runs
5
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 5 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 | 84.4 | 86.2 | 87.4 |
| Agent Workflow | 90.5 | 91.6 | 93.2 |
| Code Generation | 70.3 | 74.6 | 77.8 |
| Role Play & Narrative | 87.1 | 91.2 | 94.4 |
| Research & Analysis | 84.3 | 87.4 | 90.1 |
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 4090 | Strata | — | 72.5 tok/s | 71.2 tok/s | 80.7 GB | 262,144 tokens | 86.2 | 5 |
Benchmark runs
All 5 qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram good for coding?
- In our benchmarks, qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram scores 74.6/100 for coding. It runs at about 71.2 tok/s, so if you want more speed, unsloth/Qwen3.8-27B-GGUF:IQ3_S is faster (~104.3 tok/s) and still scores well for coding (83.0/100).
- Is qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram good for agentic (tool-using) tasks?
- In our benchmarks, qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram scores 91.6/100 for agentic workflows. It runs at about 71.2 tok/s, so if you want more speed, Qwen3.8-Flash-Next-oQ3.5e-mtp is faster (~98.5 tok/s) and still scores well for agentic workflows (92.5/100).
- How fast is qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram for local inference?
- Across 5 community benchmark runs, qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram reaches up to 72.5 tok/s and averages 71.2 tok/s, with the fastest results on NVIDIA GeForce RTX 4090.
- How much memory does qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram need?
- The leanest observed configuration used about 80.7 GB of memory.
- Which tools have been used to run qwen3.8-flash-next-ud-iq4_xs-v31-q8-engram?
- Benchmarks were submitted using Strata. Results are community-contributed and updated as new runs arrive.