Don't use answer_with_citations() for local models
c3715f2
jedick
commited on
Spaces:
Running
on
Zero
Running
on
Zero
Commit History
Enable FlashAttention
7e18a82
jedick
commited on
Change chat model to google/gemma-3-12b-it
555a40e
jedick
commited on
Use .content (avoid TypeError)
1b28bd4
jedick
commited on
Omit "Tool Call" in formatted message content
cb0db2c
jedick
commited on
Convert consecutive ToolMessages to AIMessage
b7a3bb3
jedick
commited on
Download embedding model during app startup
7320bf1
jedick
commited on
Update examples
e00a4f0
jedick
commited on
Enable thinking for answer
3575a77
jedick
commited on
Revert model download before running workflow
355c5a2
jedick
commited on
Revert "Attempt fix for ChromaDB ValueError"
5a19eb4
jedick
commited on
Attempt fix for ChromaDB ValueError
9477d3a
jedick
commited on
Further simplify ToolCallingLLM
158fae7
jedick
commited on
Improve args for thinking mode
503a0b6
jedick
commited on
Change model to Qwen3-14B
9d0646a
jedick
commited on
Lower @spaces.GPU duration
cc5c14d
jedick
commited on
Handle tool calls with thinking enabled
7d21953
jedick
commited on
Print message summaries for debugging
ba29f1f
jedick
commited on
Pin torch version
c70eaca
jedick
commited on
Download data from S3 bucket
3b06efd
jedick
commited on
Adjust prompts
6a3fed7
jedick
commited on
Rename compute_location to compute_mode
100d2c7
jedick
commited on
Normalize message types for Gemma
03db0de
jedick
commited on
Add files
08fac87
jedick
commited on