Model benchmarks
qwen3.8-27b@iq3_s local LLM performance
As of August 2026, qwen3.8-27b@iq3_s runs at up to 18.6 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
LM Studio
Model size
27B
Peak speed
18.6 tok/s
Average speed
18.4 tok/s
Min memory
12.1 GB
Max context
29,020 tokens
Best quality
94.9
Benchmark runs
2
GPUs tested
1
Performance by hardware and tool
Every hardware/tool/quantization combination qwen3.8-27b@iq3_s has been benchmarked on, ranked by peak token generation speed. Last updated August 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7600 XT | LM Studio | — | 18.6 tok/s | 18.4 tok/s | 12.1 GB | 29,020 tokens | 71.1 | 2 |
Frequently asked questions
- How fast is qwen3.8-27b@iq3_s for local inference?
- Across 2 community benchmark runs, qwen3.8-27b@iq3_s reaches up to 18.6 tok/s and averages 18.4 tok/s, with the fastest results on AMD Radeon RX 7600 XT.
- How much memory does qwen3.8-27b@iq3_s need?
- The leanest observed configuration used about 12.1 GB of memory.
- Which tools have been used to run qwen3.8-27b@iq3_s?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.