Dataset · sharegpt-vicuna
ShareGPT conversations (LMSYS collection)
Builder: LMSYS (collected from ShareGPT.com users)
Availability: never_released · checked 2026-09-24 · source
Blog: 'There is no plan to release the dataset.'
Raw record: /data/datasets/sharegpt-vicuna.json
Fields
- id
- sharegpt-vicuna
- builder
- LMSYS (collected from ShareGPT.com users)
- release_date
- nullnot_recorded
- availability
- never_released · checked 2026-09-24 · sourcenote: Blog: 'There is no plan to release the dataset.'
- content
- User-shared ChatGPT conversations from ShareGPT.com: ~70K for the first Vicuna (blog), ~125K for v1.3 (model card).recorded · source
- primary_sources
- record_history
- date:2026-09-24 · change:created from primary sources (dataset records ruling) · by:wilson-pruitt + claude ·
Parents
Influence without weights
- distilled_from_outputs → ChatGPT (Nov. 2022 launch model) declared source ShareGPT.com is a site 'where users can share their ChatGPT conversations.' User-curated, not generated by LMSYS.
Children
Training data
- ← trained_on Vicuna 7B v1.3 declared source Card: 'around 125K conversations collected from ShareGPT.com'.
- ← trained_on vicuna-7b-v1.5 declared source Same ShareGPT collection as v1.3; version doc gives 370M training tokens, same figure as v1.1/v1.3.
- ← trained_on vicuna-13b-v1.5 declared source
- ← trained_on MPT-7B-Chat declared source Archived card names 'ShareGPT-Vicuna' (link jeffwan/sharegpt_vicuna, unreachable). The `sharegpt-vicuna` record is the LMSYS collection; mapping by name. Reviewer: confirm the id mapping.
- ← trained_on StableLM-Tuned-Alpha-7B declared source Card: 'ShareGPT Vicuna (English subset)', metadata jeffwan/sharegpt_vicuna, which is now unreachable (HTTP 401), so identity with the `sharegpt-vicuna` record (LMSYS collection) is by name and could not be checked. Reviewer: confirm the id mapping.
Read in
No station on the reading path has touched this record yet.