Model benchmarks
unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL local LLM performance
As of October 2026, unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL runs at up to 69.8 tok/s for local inference (best of 4 community benchmark runs across 2 GPUs).
Model size
26B
Peak speed
69.8 tok/s
Average speed
67.1 tok/s
Avg PP
836.3 tok/s
Min memory
22.3 GB
Max context
131,072 tokens
Avg output / run
10,061 tokens
Avg runtime / run
2m 47s
Avg quality
70.3
Benchmark runs
4
GPUs tested
2
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 4 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 | 64.1 | 70.3 | 75.5 |
| Agent Workflow | 70.5 | 78.1 | 83.6 |
| Code Generation | 60.0 | 64.2 | 67.5 |
| Role Play & Narrative | 68.9 | 78.7 | 86.9 |
| Research & Analysis | 42.3 | 60.0 | 71.2 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL has been benchmarked on, ranked by peak token generation speed. Last updated October 2026.
| Hardware | Tool | Quant | Best tok/s | Avg tok/s | Memory | Context | Quality | Runs |
|---|---|---|---|---|---|---|---|---|
| AMD Radeon AI PRO R9700 | llama.cpp | Q5_K | 69.8 tok/s | 69.7 tok/s | 22.3 GB | 131,072 tokens | 75.0 | 2 |
| Advanced Micro Devices, Inc. [AMD/ATI] Device 7551 | llama.cpp | Q5_K | 64.7 tok/s | 64.5 tok/s | n/a | 65,536 tokens | 65.5 | 2 |
Benchmark runs
All 4 unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL good for coding?
- In our benchmarks, unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL scores 64.2/100 for coding. It runs at about 67.1 tok/s, so if you want more speed, qwen3.8-27b-Swift1.5-Uncensored-ATX-AI3 is faster (~85.5 tok/s) and still scores well for coding (84.8/100).
- Is unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL scores 78.1/100 for agentic workflows. It runs at about 67.1 tok/s, so if you want more speed, Nex-N2.5-mini-APEX-Mini is faster (~88.2 tok/s) and still scores well for agentic workflows (91.3/100).
- How fast is unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL for local inference?
- Across 4 community benchmark runs, unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL reaches up to 69.8 tok/s and averages 67.1 tok/s, with the fastest results on AMD Radeon AI PRO R9700.
- How much memory does unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL need?
- The leanest observed configuration used about 22.3 GB of memory (quantizations tested: Q5_K).
- Which tools have been used to run unsloth/gemma-4-26B-A4B-it-GGUF:UD-Q5_K_XL?
- Benchmarks were submitted using llama.cpp. Results are community-contributed and updated as new runs arrive.