Model benchmarks
Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K local LLM performance
As of November 2026, Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K runs at up to 43.7 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
27B
Peak speed
43.7 tok/s
Average speed
34.2 tok/s
Avg PP
471.2 tok/s
Min memory
39.8 GB
Max context
262,144 tokens
Avg output / run
51,664 tokens
Avg runtime / run
24m 46s
Avg quality
85.7
Benchmark runs
2
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 2 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 | 83.3 | 85.7 | 88.1 |
| Agent Workflow | 81.9 | 86.9 | 91.8 |
| Code Generation | 85.6 | 85.9 | 86.2 |
| Role Play & Narrative | 90.7 | 91.6 | 92.6 |
| Research & Analysis | 73.2 | 78.4 | 83.6 |
Performance by hardware and tool
Every hardware/tool/quantization combination Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K 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 RTX A6000 | llama.cpp | Q6_K | 43.7 tok/s | 34.2 tok/s | 39.8 GB | 262,144 tokens | 85.7 | 2 |
Benchmark runs
All 2 Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K good for coding?
- In our benchmarks, Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K scores 85.9/100 for coding. It runs at about 34.2 tok/s, so if you want more speed, local-qwen is faster (~60.2 tok/s) and still scores well for coding (84.8/100).
- Is Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K good for agentic (tool-using) tasks?
- In our benchmarks, Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K scores 86.9/100 for agentic workflows. It runs at about 34.2 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 Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K for local inference?
- Across 2 community benchmark runs, Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K reaches up to 43.7 tok/s and averages 34.2 tok/s, with the fastest results on NVIDIA RTX A6000.
- How much memory does Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K need?
- The leanest observed configuration used about 39.8 GB of memory (quantizations tested: Q6_K).
- Which tools have been used to run Qwen3.8-27B-Cold-Fusion-GAIN-V1.1-NM-DAU-NEO-MAX-MTP-GGUF-Q6_K?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.