Daniël de Kok 4594e6faba Add support for Marlin-quantized models ... This change adds support for Marlin-quantized models. Marlin is an FP16xINT4 matmul kernel, which provides good speedups decoding batches of 16-32 tokens. It supports quantized models with symmetric quantization, groupsize -1 or 128, and 4-bit. Tested with: - Llama 2 - Llama 3 - Phi 3		2024-06-06 13:16:52 +02:00
..
__init__.py	feat(server): flash santacoder (#153)	2023-04-03 19:06:42 +02:00
bloom_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
clip.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
flash_cohere_modeling.py	feat: move allocation logic to rust (#1835)	2024-06-05 12:18:38 +02:00
flash_dbrx_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_gemma_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_gpt2_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_llama_modeling.py	Fixing Phi3.	2024-06-01 08:47:00 +00:00
flash_mistral_modeling.py	Purely refactors paged/attention into `layers/attention` and make hardware differences more obvious with 1 file per hardware. (#1986)	2024-05-31 17:57:01 +02:00
flash_mixtral_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_neox_modeling.py	feat: move allocation logic to rust (#1835)	2024-06-05 12:18:38 +02:00
flash_pali_gemma_modeling.py	Pali gemma modeling (#1895)	2024-05-16 06:58:47 +02:00
flash_phi_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_qwen2_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_rw_modeling.py	feat: move allocation logic to rust (#1835)	2024-06-05 12:18:38 +02:00
flash_santacoder_modeling.py	Add support for Marlin-quantized models	2024-06-06 13:16:52 +02:00
flash_starcoder2_modeling.py	Purely refactors paged/attention into `layers/attention` and make hardware differences more obvious with 1 file per hardware. (#1986)	2024-05-31 17:57:01 +02:00
idefics2.py	MLPSpeculator. (#1865)	2024-05-14 12:33:18 +02:00
idefics_config.py	chore: add pre-commit (#1569)	2024-02-16 11:58:58 +01:00
idefics_image_processing.py	chore: formatting	2023-12-11 14:49:52 +01:00
idefics_modeling.py	reenable xpu for tgi (#1939)	2024-05-23 14:11:08 +02:00
idefics_perceiver.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
idefics_processing.py	chore: add pre-commit (#1569)	2024-02-16 11:58:58 +01:00
idefics_vision.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
llava_next.py	MLPSpeculator. (#1865)	2024-05-14 12:33:18 +02:00
mamba_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
mpt_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
neox_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
opt_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
phi_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
siglip.py	Removing some unused code. (#1915)	2024-05-17 11:35:49 +02:00
t5_modeling.py	Refactor layers. (#1866)	2024-05-13 12:44:30 +02:00
vlm.py	Pali gemma modeling (#1895)	2024-05-16 06:58:47 +02:00