Model benchmarks
prism-ml/Ternary-Bonsai-2-4B-gguf local LLM performance
As of October 2026, prism-ml/Ternary-Bonsai-2-4B-gguf runs at up to 155.5 tok/s for local inference (best of 2 community benchmark runs across 2 GPUs).
Model size
4B
Peak speed
155.5 tok/s
Average speed
97.0 tok/s
Avg PP
782.1 tok/s
Min memory
5.0 GB
Max context
262,144 tokens
Avg output / run
18,144 tokens
Avg runtime / run
5m 1s
Avg quality
56.1
Benchmark runs
2
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 the 1 of 2 runs with completed quality scores. 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 | — | 56.1 | — |
| Agent Workflow | — | 67.1 | — |
| Code Generation | — | 31.0 | — |
| Role Play & Narrative | — | 72.2 | — |
| Research & Analysis | — | 54.0 | — |
Performance by hardware and tool
Every hardware/tool/quantization combination prism-ml/Ternary-Bonsai-2-4B-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 | Q2_0 | 155.5 tok/s | 155.5 tok/s | 5.0 GB | 65,536 tokens | 56.1 | 1 |
| Intel(R) Arc(TM) 140V GPU (16GB) | Unsloth Studio | Q2_0 | 38.5 tok/s | 38.5 tok/s | 15.1 GB | 262,144 tokens | n/a | 1 |
Benchmark runs
All 2 prism-ml/Ternary-Bonsai-2-4B-gguf runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is prism-ml/Ternary-Bonsai-2-4B-gguf good for agentic (tool-using) tasks?
- In our benchmarks, prism-ml/Ternary-Bonsai-2-4B-gguf scores 67.1/100 for agentic workflows. It runs at about 97.0 tok/s, so if you want more speed, cyber-tiel-coder-35b-a3b-mtp is faster (~115.9 tok/s) and still scores well for agentic workflows (86.4/100).
- How fast is prism-ml/Ternary-Bonsai-2-4B-gguf for local inference?
- Across 2 community benchmark runs, prism-ml/Ternary-Bonsai-2-4B-gguf reaches up to 155.5 tok/s and averages 97.0 tok/s, with the fastest results on NVIDIA GeForce RTX 4070.
- How much memory does prism-ml/Ternary-Bonsai-2-4B-gguf need?
- The leanest observed configuration used about 5.0 GB of memory (quantizations tested: Q2_0).
- Which tools have been used to run prism-ml/Ternary-Bonsai-2-4B-gguf?
- Benchmarks were submitted using Unsloth Studio. Results are community-contributed and updated as new runs arrive.