Model benchmarks
ukisai/Swift-1.5-Qwen3.8-27B-GGUF local LLM performance
As of September 2026, ukisai/Swift-1.5-Qwen3.8-27B-GGUF runs at up to 64.7 tok/s for local inference (best of 3 community benchmark runs across 1 GPU).
Model size
27B
Peak speed
64.7 tok/s
Average speed
55.3 tok/s
Avg PP
186.3 tok/s
Min memory
1164.8 GB
Max context
131,072 tokens
Avg output / run
26,940 tokens
Avg runtime / run
8m 45s
Avg quality
78.1
Benchmark runs
3
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 3 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 | 69.0 | 78.1 | 87.7 |
| Agent Workflow | 85.6 | 87.3 | 90.0 |
| Code Generation | 15.3 | 52.4 | 80.0 |
| Role Play & Narrative | 77.4 | 88.0 | 95.0 |
| Research & Analysis | 82.7 | 84.8 | 87.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination ukisai/Swift-1.5-Qwen3.8-27B-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 |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon RX 7900 XT/7900 XTX/7900M | Unsloth Studio | Q6_K | 64.7 tok/s | 62.7 tok/s | 1449.2 GB | 131,072 tokens | 83.2 | 2 |
| AMD Radeon RX 7900 XT/7900 XTX/7900M | Unsloth Studio | Q4_K_M | 40.7 tok/s | 40.7 tok/s | 1164.8 GB | 8,192 tokens | 68.1 | 1 |
Benchmark runs
All 3 ukisai/Swift-1.5-Qwen3.8-27B-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is ukisai/Swift-1.5-Qwen3.8-27B-GGUF good for coding?
- In our benchmarks, ukisai/Swift-1.5-Qwen3.8-27B-GGUF scores 52.4/100 for coding. It runs at about 55.3 tok/s, so if you want more speed, ornith-ai/Ornith-1.5-35B-A3B-GGUF is faster (~96.3 tok/s) and still scores well for coding (83.4/100).
- Is ukisai/Swift-1.5-Qwen3.8-27B-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, ukisai/Swift-1.5-Qwen3.8-27B-GGUF scores 87.3/100 for agentic workflows. It runs at about 55.3 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 ukisai/Swift-1.5-Qwen3.8-27B-GGUF for local inference?
- Across 3 community benchmark runs, ukisai/Swift-1.5-Qwen3.8-27B-GGUF reaches up to 64.7 tok/s and averages 55.3 tok/s, with the fastest results on AMD Radeon RX 7900 XT/7900 XTX/7900M.
- How much memory does ukisai/Swift-1.5-Qwen3.8-27B-GGUF need?
- The leanest observed configuration used about 1164.8 GB of memory (quantizations tested: Q4_K_M, Q6_K).
- Which tools have been used to run ukisai/Swift-1.5-Qwen3.8-27B-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.