Spaces:
Sleeping
Sleeping
Commit History
-
Pass HF_TOKEN to embedding model to avoid 429 rate limits
vaishnav +1: Claude Opus 4.6 (1M context)Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>93c3adb -
Stream LLM tokens in real-time, trim history, and reduce retriever k
vaishnav +1: Claude Opus 4.6 (1M context)- Replace fake chunked streaming with real .stream() from the RAG chain so users see tokens as they arrive from the LLM - Trim chat history to last 10 messages to bound prompt size - Drop retriever k from 100 to 10 for faster, cheaper inference Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>7abe457
-
Restore full URL list in configs/config.py
vaishnav +1: Claude Opus 4.6 (1M context)Previous commit accidentally left the URL list mostly commented out from local testing. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>1f71c2d -
Raise HF max_new_tokens and load .env from any cwd
vaishnav +1: Claude Opus 4.6 (1M context)- Bump max_new_tokens to 4096 (default 512 was truncating answers) - Use find_dotenv(usecwd=True) so launching from any directory works Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>51f4eed -
Switch HF default to openai/gpt-oss-120b with auto provider routing
vaishnav +1: Claude Opus 4.6 (1M context)meta-llama/Llama-3.1-8B-Instruct returned 404 on hf-inference. Drop the explicit provider= so HuggingFaceEndpoint / InferenceClient picks a working route automatically, and default to openai/gpt-oss-120b. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>1c2dc4f
-
Unpin gradio to match HF Spaces base image
vaishnav +1: Claude Opus 4.6 (1M context)HF Spaces installs gradio==5.23.2 on top of requirements.txt, which conflicted with the pinned 5.18.0 and broke the build. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>d19ec61 -
Fix HuggingFace provider crash and harden chain init
vaishnav +1: Claude Opus 4.6 (1M context)- Pass provider="hf-inference" and task="conversational" to HuggingFaceEndpoint to avoid huggingface_hub StopIteration -> RuntimeError on empty provider mapping - Switch default model to meta-llama/Llama-3.1-8B-Instruct (served by hf-inference) - Make set_envs() a no-op so non-interactive deploys don't hang on getpass - Add LFUCache.delete() for session-scoped history clearing - Update CLAUDE.md to reflect multi-provider setup and constraints Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>332bfab
-
change embeddings
vaishnav8153c79 -
Update configs/config.py
Verified signatureb2790bb -
update requirements
vaishnav4a14f2d -
update requirements
vaishnav93f5b6b -
update requirements
vaishnav79c10e9 -
update requirements
vaishnav6debb69 -
update requirements.txt
vaishnav9b7f3b8
-
make mistral default model
vaishnava856301 -
refactor UI and embedding of URL's
vaishnavbcf9d83
-
pull latest changes
c9c38b4 -
Add CLAUDE.md for Claude Code guidance
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>81f19ca -
Update requirements.txt
Verified signature18c9d48
-
Update README.md
Verified signature107cde7
-
Update requirements.txt
Verified signature7f7a66d -
Update README.md
Verified signature460db29 -
Update README.md
Verified signature3db51e9
-
Update configs/config.py
Verified signature185beff
-
Update configs/config.py
Verified signature6170811 -
Update app.py
Verified signaturefd7a2e7 -
Update configs/config.py
Verified signatureupdate SYSTEM_PROMPTa9765a6 -
Update llm_setup/llm_setup.py
Verified signatureRefactor context prompt7dbb9b2 -
Update configs/config.py
Verified signature0f57fd9 -
Update configs/config.py
Verified signatureAdd more URL's to knowledge base81dc6ba
-
Update configs/config.py
Verified signatureb84a5b8 -
Update configs/config.py
Verified signature7c9764c
-
Update configs/config.py
Verified signatureb32ad4a
-
Update configs/config.py
Verified signatureuse gemini-2.5-pro-exp-03-252774314 -
Update configs/config.py
Verified signatureuse gemini-2.0-flash-lite4b22966
-
Create .env
Verified signature164c4cf -
Update README.md
Verified signaturef8aafda
-
Update README.md
Verified signature51af172
-
Update configs/config.py
Verified signaturee58dfbf -
Update configs/config.py
Verified signatureca4b8f3 -
Create .env
Verified signature9b1df6c
-
make it a framework
vaishnav429af6c -
implement streaming
vaishnavf2cbb22
-
Update configs/config.py
Verified signaturefaa8f77 -
Update configs/config.py
Verified signaturef50c60a -
Update configs/config.py
Verified signaturec4f1997 -
Add prompt to ask followup questions
Verified signaturea7d936d -
Update README.md
Verified signature4969d0b -
Update configs/config.py
Verified signature498d79e
-
Merge branch 'main' of hf.co:spaces/vaishnaveswar/AIVIZ-BOT
vaishnav59bfa04