Model benchmarks
unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF local LLM performance
As of October 2026, unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF runs at up to 133.9 tok/s for local inference (best of 4 community benchmark runs across 1 GPU).
Model size
4B
Peak speed
133.9 tok/s
Average speed
121.1 tok/s
Avg PP
3301.0 tok/s
Min memory
12.0 GB
Max context
262,144 tokens
Avg output / run
7,808 tokens
Avg runtime / run
1m 6s
Avg quality
69.0
Benchmark runs
4
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 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 | 64.8 | 69.0 | 71.9 |
| Agent Workflow | 81.7 | 85.1 | 88.9 |
| Code Generation | 51.0 | 57.8 | 63.8 |
| Role Play & Narrative | 63.6 | 70.3 | 75.4 |
| Research & Analysis | 50.1 | 62.7 | 71.5 |
Performance by hardware and tool
Every hardware/tool/quantization combination unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF 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 |
|---|---|---|---|---|---|---|---|---|
| NVIDIA GeForce RTX 4070 | Unsloth Studio | UD-Q4_K_XL | 133.9 tok/s | 121.1 tok/s | 12.0 GB | 262,144 tokens | 69.0 | 4 |
Benchmark runs
All 4 unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF good for agentic (tool-using) tasks?
- In our benchmarks, unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF scores 85.1/100 for agentic workflows. It runs at about 121.1 tok/s, so if you want more speed, ornith-1.5:35b is faster (~151.9 tok/s) and still scores well for agentic workflows (90.7/100).
- How fast is unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF for local inference?
- Across 4 community benchmark runs, unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF reaches up to 133.9 tok/s and averages 121.1 tok/s, with the fastest results on NVIDIA GeForce RTX 4070.
- How much memory does unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF need?
- The leanest observed configuration used about 12.0 GB of memory (quantizations tested: UD-Q4_K_XL).
- Which tools have been used to run unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.