| .. |
|
__init__.py
|
bbde979ecd
DeepSeek-V2 (#579)
|
hace 7 meses |
|
arctic.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
baichuan.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
bloom.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
chatglm.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
clip.py
|
3a0fdf7b9b
chore: remove `image_input_type` from VLM config
|
hace 7 meses |
|
commandr.py
|
da6765c084
feat: lora support for commandr models
|
hace 7 meses |
|
dbrx.py
|
b2cb5a92e9
fix: missing cache_config for dbrx
|
hace 7 meses |
|
decilm.py
|
56e0b8223c
chore: add base class for LoRA-supported models
|
hace 7 meses |
|
deepseek.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
deepseek_v2.py
|
bbde979ecd
DeepSeek-V2 (#579)
|
hace 7 meses |
|
falcon.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
gemma.py
|
b6ff0623a6
chore: clean up branding
|
hace 7 meses |
|
gpt2.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
gpt_bigcode.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
gpt_j.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
gpt_neox.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
interfaces.py
|
85ef2fe8b1
chore: clean up placeholder symbols
|
hace 7 meses |
|
internlm2.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
jais.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
llama.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
llama_embedding.py
|
50b7c13db0
refactor: attention selector (#552)
|
hace 7 meses |
|
llava.py
|
3a0fdf7b9b
chore: remove `image_input_type` from VLM config
|
hace 7 meses |
|
llava_next.py
|
3a0fdf7b9b
chore: remove `image_input_type` from VLM config
|
hace 7 meses |
|
minicpm.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
mixtral.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
mixtral_quant.py
|
b6ff0623a6
chore: clean up branding
|
hace 7 meses |
|
mlp_speculator.py
|
de7e6919c0
feat: support tied weights and input scale for MLPSpeculator
|
hace 7 meses |
|
mpt.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
olmo.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
opt.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
orion.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
phi.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
phi3_small.py
|
696f2cd59c
add phi3_small support with blocksparse attention
|
hace 7 meses |
|
phi3v.py
|
3a0fdf7b9b
chore: remove `image_input_type` from VLM config
|
hace 7 meses |
|
qwen.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
qwen2.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
|
qwen2_moe.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
stablelm.py
|
656459fd84
make fp8_e4m3 work on nvidia
|
hace 7 meses |
|
starcoder2.py
|
ac79d115b3
add guards for prefix caching, fp8, chunked, etc
|
hace 7 meses |
|
xverse.py
|
c5d8028668
fix: no need to redefine supports_vision and supports_lora in model class
|
hace 7 meses |
AlpinDale
3a0fdf7b9b
chore: remove `image_input_type` from VLM config