Model benchmarks
bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF local LLM performance
As of September 2026, bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF runs at up to 44.1 tok/s for local inference (best of 5 community benchmark runs across 1 GPU).
Model size
27B
Peak speed
44.1 tok/s
Average speed
37.8 tok/s
Avg PP
410.4 tok/s
Min memory
24.4 GB
Max context
80.000 tokens
Avg output / run
49.049 tokens
Avg runtime / run
21m 24s
Avg quality
85.0
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 | 83.7 | 85.0 | 86.3 |
| Agent Workflow | 76.2 | 82.0 | 88.5 |
| Code Generation | 78.4 | 80.9 | 82.7 |
| Role Play & Narrative | 87.7 | 92.1 | 94.3 |
| Research & Analysis | 84.7 | 85.1 | 85.9 |
Performance by hardware and tool
Every hardware/tool/quantization combination bottlecapai/ThinkingCap-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 XTX | Unsloth Studio | Q4_K_M | 44.1 tok/s | 37.8 tok/s | 24.4 GB | 80.000 tokens | 85.0 | 5 |
Benchmark runs
All 5 bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF good for coding?
- In our benchmarks, bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF scores 80.9/100 for coding. It runs at about 37.8 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 bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF scores 82.0/100 for agentic workflows. It runs at about 37.8 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 bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF for local inference?
- Across 5 community benchmark runs, bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF reaches up to 44.1 tok/s and averages 37.8 tok/s, with the fastest results on AMD Radeon RX 7900 XTX.
- How much memory does bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF need?
- The leanest observed configuration used about 24.4 GB of memory (quantizations tested: Q4_K_M).
- Which tools have been used to run bottlecapai/ThinkingCap-Qwen3.8-27B-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.