Awni Hannun 5cae0a60e6 deepseek v3 model with pipeline parallelism (#1191) ...

* deepseekv3

* use upload_large_file instead of deprecated multi comit

* add pipeline generation and example

* comment

* get fp16 working

* use mlx==0.22

2025-01-09 15:55:53 -08:00

..

test_datsets.py

fix encoding with special tokens + chat template (#1189)

2025-01-03 10:50:59 -08:00

test_finetune.py

Enable distributed LoRA training (#821)

2024-11-02 18:02:31 -07:00

test_generate.py

Generation refactor: part 2 (#1099)

2024-11-23 11:47:06 -08:00

test_gguf.py

fix(mlx-lm): type hints in gguf.py (#621)

2024-03-26 07:56:01 -07:00

test_models.py

deepseek v3 model with pipeline parallelism (#1191)

2025-01-09 15:55:53 -08:00

test_prompt_cache.py

Allow prompt callback to generate_step (#1133)

2024-12-03 16:17:14 -08:00

test_sample_utils.py

Fix no template prompt + top_k sampling (#1166)

2024-12-18 18:46:50 -08:00

test_server.py

Prompt caching in mlx_lm.server (#1026)

2024-10-14 10:57:22 -07:00

test_tokenizers.py

Fix decoding manually added tokens (#1164)

2024-12-17 09:54:29 -08:00

test_tuner_utils.py

LoRA: Extract small function (#614)

2024-06-02 06:38:42 -07:00

test_utils_load_model.py

deepseek v3 model with pipeline parallelism (#1191)

2025-01-09 15:55:53 -08:00

test_utils.py

Fix whipser conversion for safetensors models (#935)

2024-08-14 10:22:04 -07:00