tag:github.com,2008:https://github.com/mozilla-ai/structured-qa/releases 2025-02-06T16:15:28Z tag:github.com,2008:Repository/904169776/1.0.4 2025-02-06T16:16:26Z <p>Update default model to `bartowski/Qwen2.5-7B-Instruct-GGUF/Qwen2.5-7…</p> <p>…B-Instruct-Q8_0.gguf`</p> daavoo tag:github.com,2008:Repository/904169776/1.0.3 2025-02-06T15:54:37Z daavoo tag:github.com,2008:Repository/904169776/1.0.2 2025-02-06T14:53:30Z <p>Merge pull request <a class="issue-link js-issue-link" href="https://github.com/mozilla-ai/structured-qa/pull/18">#18</a> from mozilla-ai/17-cli-run-bug</p> <p>fix(cli): Load `question` from config.</p> daavoo tag:github.com,2008:Repository/904169776/1.0.1 2025-02-05T16:43:18Z daavoo tag:github.com,2008:Repository/904169776/1.0.0 2025-02-05T16:22:15Z <p>5 add benchmark (<a class="issue-link js-issue-link" href="https://github.com/mozilla-ai/structured-qa/pull/16">#16</a>)</p> <p>* enh(preprocessing): Add split_markdown_by_headings.</p> <p>* Add benchmark</p> <p>* Move to structured_qa. Add entrypoint</p> <p>* Move back outside</p> <p>* Fix main</p> <p>* Update questions</p> <p>* Update model and prompt</p> <p>* Update</p> <p>* Update</p> <p>* fix</p> <p>* Add system_instruction</p> <p>* Update ratio</p> <p>* Add more wait</p> <p>* Fix return</p> <p>* Fix URLs</p> <p>* Update download name</p> <p>* Update</p> <p>* Update</p> <p>* Update with upper</p> <p>* Cast to str</p> <p>* Extend</p> <p>* Add benchmark</p> <p>* Fix</p> <p>* fix</p> <p>* Drop export</p> <p>* Updates</p> <p>* Update default model</p> <p>* Update</p> <p>* Use info</p> <p>* Update with None</p> <p>* Add answer type</p> <p>* Refactor</p> <p>* Add fallback for out of context</p> <p>* Update with debugging info</p> <p>* Update</p> <p>* Update with mit-1</p> <p>* test unsloth</p> <p>* Add , skip_special_tokens = True</p> <p>* Update</p> <p>* Updates</p> <p>* Add full_context</p> <p>* Update full context</p> <p>* update</p> <p>* Add load and clean</p> <p>* Update</p> <p>* Update</p> <p>* print</p> <p>* Update</p> <p>* Add load_gemini_model</p> <p>* Add sleep</p> <p>* Update get_response</p> <p>* Update</p> <p>* Log error</p> <p>* fix</p> <p>* Make the more info check more flexible</p> <p>* Add gemini_full_context notebook</p> <p>* typo</p> <p>* Check por API KEY</p> <p>* Update with outputs</p> <p>* Add ragatouille</p> <p>* Fix</p> <p>* Update notebooks</p> <p>* Update gemini notebooks</p> <p>* Extend structured_qa. Add perfect_context.</p> <p>* Add gemini_perfect_context</p> <p>* Update</p> <p>* fix line</p> <p>* fix line</p> <p>* Update perfect_context</p> <p>* Add missing perfect context</p> <p>* Updates</p> <p>* Update gemini_ragatouille</p> <p>* Update gemini_fra</p> <p>* Update</p> <p>* Update</p> <p>* Drop some log</p> <p>* Update</p> <p>* Update gemini_perfect_context with results</p> <p>* Use rapizfuzz</p> <p>* Use question_part</p> <p>* Fix</p> <p>* break when no section_names</p> <p>* Update prompt</p> <p>* Add qwen perfect context</p> <p>* Update gemini_find_retrieve_answer</p> <p>* Update qwen perfect context</p> <p>* Add qwen RAGatouille</p> <p>* Update qwen notebooks</p> <p>* Update</p> <p>* Update prompt</p> <p>* Update qwen notebooks</p> <p>* Cleanup</p> <p>* Cleanup</p> <p>* Add DeepSeek-R1-Distill-Qwen-7B</p> <p>* Debug current calls. Set to 9 before reset</p> <p>* Add qwen find retrieve answer</p> <p>* Extend benchmark</p> <p>* Update</p> <p>* Add max_sections_to_check</p> <p>* Default to None</p> <p>* Default to half of sections</p> <p>* Update</p> <p>* fix</p> <p>* Fix</p> <p>* Add qwen full context</p> <p>* Update qwen_full_context</p> <p>* Update gemini_full_context</p> <p>* Add statistics</p> <p>* Update prompt</p> <p>* Update with type</p> <p>* Update gemini prompt and count</p> <p>* Update results with same prompts</p> <p>* Update with same prompt</p> <p>* Update results</p> <p>* Bring back llama-cpp-python</p> <p>* Update prompts</p> <p>* Reduce notebook size</p> <p>* Update pre-commit</p> <p>* Update docstrings</p> <p>* Update test</p> daavoo tag:github.com,2008:Repository/904169776/0.3.2 2025-02-04T13:34:43Z <p>enh(app): Update default model to bartowski/Qwen2.5-3B-Instruct-GGUF/…</p> <p>…Qwen2.5-3B-Instruct-f16.gguf</p> daavoo tag:github.com,2008:Repository/904169776/0.3.1 2025-01-17T11:43:37Z daavoo tag:github.com,2008:Repository/904169776/0.3.0 2025-01-17T11:29:16Z <p>Add Dockerfile and sync_hf_space (<a class="issue-link js-issue-link" href="https://github.com/mozilla-ai/structured-qa/pull/12">#12</a>)</p> <p>* Add Dockerfile and sync_hf_space</p> <p>* Add reboot_space</p> <p>* Add README.md</p> <p>* Check GPU without torch</p> <p>* Update</p> daavoo tag:github.com,2008:Repository/904169776/0.2.0 2025-01-16T15:33:04Z <p>6 add docs (<a class="issue-link js-issue-link" href="https://github.com/mozilla-ai/structured-qa/pull/8">#8</a>)</p> <p>* Update API</p> <p>* Update index</p> <p>* Fix repo_url</p> <p>* Update getting-started</p> <p>* Updates with model loader</p> <p>* Update step-by-step guide</p> <p>* Update defaults</p> <p>* Update contributions</p> <p>* reorder</p> <p>* Lint</p> <p>* Fix import</p> <p>* fix test_workflow</p> <p>* Update README.md</p> <p>* Update site_name</p> <p>* Apply suggestions from code review</p> daavoo tag:github.com,2008:Repository/904169776/0.1.0 2025-01-15T17:10:38Z daavoo