SALT-NLP/SWE-chat
SWE-chat: Coding Agent Interactions From Real Users in the Wild ๐ COLM 2026 Paper: arxiv.org/abs/2604.20779 ๐ Website: swe-chat.com Dataset Summary SWE-chat captures real-world AI coding sessions from developers using AI coding assistants (Claude Code, Codex, Gemini CLI, and others via the Entire.io CLI). Each session includes the full conversation transcript, tool calls, thinking traces, code changes, and attribution of human vs. agent-authored code.โฆ See the full description on the dataset page: https://huggingface.co/datasets/SALT-NLP/SWE-chat.
| Author / organization | SALT-NLP |
|---|---|
| License (as declared) | odc-by |
| Downloads | 12.2K (as reported by Hugging Face) |
| Likes | 130 |
| Created | 24 April 2026 |
| Last updated at source | 3 October 2026 |
Popularity, last 3 days
12,249 now (+343 since 2026-10-15). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
Related papers
More from SALT-NLP
- SALT-NLP/hleDataset
- SALT-NLP/cogym-collabskill-trajectoriesDataset
- SALT-NLP/cogym-real-trajectoriesDataset
- SALT-NLP/hle-tgtDataset
- SALT-NLP/Design2CodeDataset
- SALT-NLP/Sketch2CodeDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.