FrogNano-4B
Microsoft llmResearch repository-level coding agent RL-trained from Qwen3.5-4B on ~1,500 synthetic SWE tasks.
overall ⌄
32.8
112 of 125 ranked
price · $/M tokens
self-hosted
open weights, no first-party API
to run it yourself ⌄
2 GB
minimum · 10 GB recommended
Benchmarks
What its maker published, and what anyone else measured. Every figure links the document it came from.
| benchmark | published | measured |
|---|---|---|
| SWE-bench | 61.5 | — |
| SWE-bench Pro | 37.6 | — |
Lineage
No recorded lineage — root or standalone model.
Metadata
| org | Microsoft |
| released | 2026-09-17 |
| params | 4B |
| license | mit |
| hf id | microsoft/FrogNano-4B-2609 |