Patrick Devine
|
5f74d1fd47
gemma2 impl
|
2 months ago |
Jesse Gross
|
a1cda80bcb
model: Update encoder cache to use multimodal input processing handler
|
1 month ago |
Michael Yang
|
7bae7fa5ce
ml/backend/ggml: create tensor on specific backend
|
2 months ago |
Michael Yang
|
bab6f34dc0
ml/backend/ggml: update model loading for hybrid/multi backends
|
2 months ago |
Daniel Hiltgen
|
1fdb351c37
New engine: vision models and auto-fallback (#9113)
|
2 months ago |
Jesse Gross
|
854a9195f3
attention: Remove unnecessary contiguous operations
|
2 months ago |
Michael Yang
|
53d2990d9b
model: add bos token if configured
|
2 months ago |
Jesse Gross
|
f53f4198c3
ml: Abstract attention out of model definitions
|
2 months ago |
Jesse Gross
|
5c5535c064
models: Prune unused outputs earlier in the forward pass
|
2 months ago |
Jesse Gross
|
ed443a0393
Runner for Ollama engine
|
4 months ago |
Jesse Gross
|
6945617af5
models: Move model into their own directory
|
2 months ago |