v0.1.4
oumi-ai/oumiv0.1.4Feb 3, 2025by oelachqar
AI Summary
This release adds DeepSeek R1 configurations, updates VLM notebooks, and fixes the Qwen2-VL evaluation job.
Key Highlights
- Added Deepseek R1 1.5B/32B configs
- Added multi-modal VLM notebook with Llama 11B
- Fixed Qwen2-VL evaluation job to pass
- Updated documentation links and added Apache license headers
New Features
- DeepSeek R1 inference configs (Together AI)
- VLM Multimodal notebook updates
- Phi3 support for multiple images
- Support for `dataset_name_override` dataset_kwarg
Full Release Notes
## What's Changed * Add memory cleanup calls in e2e integration tests by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1277 * Set up versioning for our documentation by @taenin in https://github.com/oumi-ai/oumi/pull/1275 * Make `qwen2-VL` evaluation job pass by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1278 * Add multi-modal (vlm) notebook with Llama 11B by @optas in https://github.com/oumi-ai/oumi/pull/1258 * Documentation: Inference -> List supported models by @kaisopos in https://github.com/oumi-ai/oumi/pull/1279 * [tiny] update website link by @oelachqar in https://github.com/oumi-ai/oumi/pull/1280 * Update all documentation links to the new doc URL by @taenin in https://github.com/oumi-ai/oumi/pull/1281 * Update Oumi - A Tour.ipynb by @brragorn in https://github.com/oumi-ai/oumi/pull/1282 * Documentation: Judge (minor edits) by @kaisopos in https://github.com/oumi-ai/oumi/pull/1283 * Fix citation by @oelachqar in https://github.com/oumi-ai/oumi/pull/1285 * Add Deepseek R1 1.5B/32B configs by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1276 * Misc eval configs cleanup by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1286 * [docs] Describe parallel evaluation by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1284 * Update `microsoft/Phi-3-vision-128k-instruct` training config by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1287 * Add Together Deepseek R1 inference config by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1289 * [minor] vlm notebook minor updates (doc referencing, freeze visual backbone) by @optas in https://github.com/oumi-ai/oumi/pull/1288 * Add missing `-m oumi evaluate` argument in eval config by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1291 * [docs] Add more references to VL-SFT and SFT notebooks by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1293 * Eval config change for `deepseek-ai/DeepSeek-R1-Distill-Llama-70B` by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1292 * [notebooks] Update intro & installation instruction by @oelachqar in https://github.com/oumi-ai/oumi/pull/1294 * Update notebook intros by @oelachqar in https://github.com/oumi-ai/oumi/pull/1296 * [notebooks] Update installation instructions for colab by @oelachqar in https://github.com/oumi-ai/oumi/pull/1297 * Add Apache license header to `src/oumi/**/*.py` by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1290 * Minor updates to VLM Multimodal notebook by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1299 * [docs] Add latest notebooks and update references by @oelachqar in https://github.com/oumi-ai/oumi/pull/1300 * [tiny] Add docs auto-generated `.rst` files to gitignore by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1298 * [tiny] use GitHub link for header by @oelachqar in https://github.com/oumi-ai/oumi/pull/1301 * [docs][tiny] update inference engines reference by @oelachqar in https://github.com/oumi-ai/oumi/pull/1302 * Update README/docs to add new DeepSeek models by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1304 * [docs] Use `pip install oumi` over `pip install .` by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1305 * Tune VLM SFT configs by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1306 * Tune VLM configs for SmolVLM and Qwen2-VL by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1307 * Update config/notebook pip installs to use PyPI by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1308 * [tiny] upgrade torch version by @oelachqar in https://github.com/oumi-ai/oumi/pull/1295 * Update logging and unit tests related to chat templates by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1311 * fix(docs): "interested by joining" to "interested in joining" by @CharlesCNorton in https://github.com/oumi-ai/oumi/pull/1312 * Add HF_TOKEN instructions to Oumi Multimodal notebook by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1313 * Update configuration.md by @penfever in https://github.com/oumi-ai/oumi/pull/1314 * remove duplicate keys in config example by @lucyknada in https://github.com/oumi-ai/oumi/pull/1315 * [Notebooks] Update VLM notebook by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1317 * Update parasail_inference_engine.py by @jgreer013 in https://github.com/oumi-ai/oumi/pull/1320 * Fix typo and update warning message for OUMI trainer by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1319 * [Notebooks] Add a note that a notebook kernel restart may be needed after `pip install oumi` by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1318 * Update Phi3 to support multiple images by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1321 * Add more detailed comment headers to YAML configs by @wizeng23 in https://github.com/oumi-ai/oumi/pull/1310 * [Notebooks] Add a note to Tour notebook to restart kernel after the first `pip install` by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1327 * Tweak `--mem-fraction-static` param in sample SGLang configs by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1328 * Disallow using `DatasetParams` field names as keys in `DatasetParams.dataset_kwargs` by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1324 * Support `dataset_name_override` dataset_kwarg by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1188 * Add an util and a test marker for HF token by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1329 * Update `llama3-instruct` chat template to align with the original models template by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1326 * chore: update launcher.sh by @eltociear in https://github.com/oumi-ai/oumi/pull/1333 * [Notebooks] Minor improvements in VLM and CNN notebooks by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1335 * Update VLM cluster names in sample commands by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1336 * Update our README and docs with the github trending badge. by @taenin in https://github.com/oumi-ai/oumi/pull/1340 * Update README.md - Add DeepSeek to supported models by @mkoukoumidis in https://github.com/oumi-ai/oumi/pull/1343 * Update index.md - Add DeepSeek to supported models by @mkoukoumidis in https://github.com/oumi-ai/oumi/pull/1344 * Update "GPU Tests" status badge in README page by @xrdaukar in https://github.com/oumi-ai/oumi/pull/1345 ## New Contributors * @CharlesCNorton made their first contribution in https://github.com/oumi-ai/oumi/pull/1312 * @lucyknada made their first contribution in https://github.com/oumi-ai/oumi/pull/1315 * @eltociear made their first contribution in https://github.com/oumi-ai/oumi/pull/1333 **Full Changelog**: https://github.com/oumi-ai/oumi/compare/v0.1.3...v0.1.4