Model benchmarks
croissantllmchat-v0.1 local LLM performance
As of October 2026, croissantllmchat-v0.1 runs at up to 65.5 tok/s for local inference (best of 2 community benchmark runs across 1 GPU).
Model size
Unknown
Peak speed
65.5 tok/s
Average speed
65.2 tok/s
Avg PP
2478.2 tok/s
Min memory
2.1 GB
Max context
16,384 tokens
Avg output / run
1,515 tokens
Avg runtime / run
31s
Avg quality
7.1
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 | 5.1 | 7.1 | 9.0 |
| Agent Workflow | 5.2 | 11.3 | 17.3 |
| Code Generation | 4.5 | 5.5 | 6.4 |
| Role Play & Narrative | 3.3 | 6.0 | 8.8 |
| Research & Analysis | 3.6 | 5.6 | 7.5 |
Performance by hardware and tool
Every hardware/tool/quantization combination croissantllmchat-v0.1 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 2050 | LM Studio | Q8_0 | 65.5 tok/s | 65.2 tok/s | 2.1 GB | 16,384 tokens | 7.1 | 2 |
Benchmark runs
All 2 croissantllmchat-v0.1 runs submitted so far — expand one for its hardware, quality breakdown and per-scenario detail.
Frequently asked questions
- Is croissantllmchat-v0.1 good for coding?
- In our benchmarks, croissantllmchat-v0.1 scores 5.5/100 for coding. It runs at about 65.2 tok/s, so if you want more speed, nvidia/Qwen3.8-27B-NVFP4 is faster (~75.5 tok/s) and still scores well for coding (83.5/100).
- Is croissantllmchat-v0.1 good for agentic (tool-using) tasks?
- In our benchmarks, croissantllmchat-v0.1 scores 11.3/100 for agentic workflows. It runs at about 65.2 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 croissantllmchat-v0.1 for local inference?
- Across 2 community benchmark runs, croissantllmchat-v0.1 reaches up to 65.5 tok/s and averages 65.2 tok/s, with the fastest results on NVIDIA GeForce RTX 2050.
- How much memory does croissantllmchat-v0.1 need?
- The leanest observed configuration used about 2.1 GB of memory (quantizations tested: Q8_0).
- Which tools have been used to run croissantllmchat-v0.1?
- Benchmarks were submitted using LM Studio. Results are community-contributed and updated as new runs arrive.