ml-explore mlx-examples issues

ml-explore / mlx-examples

Examples in the MLX framework

MIT License

6.3k stars 898 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Getting TypeError: 'module' object is not subscriptable on generations with prompts over a certain size

#1022 chimezie closed 1 month ago
0
T5 tokenizer decoding error with CodeT5+

#1021 zcbenz opened 1 month ago
1
MusicGen

#1020 barronalex closed 1 month ago
6
[BUG] generate hangs after multiple iterations

#1019 hschaeufler closed 1 month ago
3
`mlx_lm.chat` command is not available after installing `mlx-lm`

#1018 gh640 closed 1 month ago
3
[BUG] Different Default-Values for Temperature

#1017 hschaeufler closed 1 month ago
1
Return logprobs from generate api

#1016 farris closed 1 month ago
2
More cache improvements

#1015 awni closed 1 month ago
9
Fix rotating KV cache for chat use case

#1014 awni closed 1 month ago
1
Make T5 work with official models without conversions

#1013 zcbenz closed 1 month ago
2
mlx_whisper: add support for audio input from stdin

#1012 anthonywu closed 3 weeks ago
2
libc++abi: terminating due to uncaught exception of type std::runtime_error: GPU FFT is only implemented for 1D, forward, complex FFTs.

#1011 CrispStrobe closed 2 months ago
3
adding-support-for-mamba2

#1009 Goekdeniz-Guelmez opened 2 months ago
8
Support for Nvidia Nemotron and NVLM 1.0

#1007 vlbosch opened 2 months ago
1
Allow conversion of Whisper turbo models

#1006 awni closed 2 months ago
0
Llama 3.2 1b model not generating response

#1005 mustangs0786 closed 2 months ago
3
repetiton_penalty and logits_bias just using logits_processors

#1004 nathanrchn closed 2 months ago
2
Server: support function calling

#1003 madroidmaq closed 2 months ago
5
EOS Token Warning

#1002 mike-schiller closed 2 months ago
2
PR: Add KV-cache creation capability to mlx_lm.generate for after a text completion

#1001 mark-lord closed 1 month ago
3
RotatingKVCache: Problem when reusing cache between multiple generations

#1000 zcbenz closed 1 month ago
1
“tokenizer.ggml.unknown_token_id” is it definitly required?

#999 fynv closed 2 months ago
2
Fix llava model when using text-only prompt

#998 zcbenz closed 2 months ago
0
LoRA: Support HuggingFace dataset via data parameter

#996 madroidmaq closed 2 months ago
0
LoRA: support tools(function calling) format datasets

#995 madroidmaq closed 2 months ago
5
ModuleNotFoundError: No module named 'huggingface_hub.utils._errors'

#994 ndurner closed 2 months ago
3
Fix export to gguf

#993 angeloskath closed 2 months ago
0
Encodec

#991 awni closed 2 months ago
1
[BUG] mlx_lm.fuse aborts with error message "AttributeError: 'bytes' object has no attribute 'encode'."

#992 hschaeufler closed 2 months ago
3
Don't use private exception class (not in latest huggingface_hub)

#990 awni closed 2 months ago
0
Enable caching for 'generate' and 'stream_generate' functions to ensure persistence of cache across multiple requests

#989 nath1295 closed 1 month ago
3
huggingface_hub 0.25 causes error upon importing from mlx_lm

#988 arogister closed 2 months ago
1
Backward Hooks

#987 Aakriti23 closed 2 months ago
3
Adding an equivalent pytorch or tensorflow of the models for true benchmark

#986 thegodone closed 1 month ago
1
Learning rate approaches warmup_init value

#985 hschaeufler closed 2 months ago
2
Add /v1/models endpoint to mlx_lm.server

#984 jamesm131 closed 2 months ago
3
Add logits_processor option to generate_step function

#983 nathanrchn closed 2 months ago
6
details on scale parameter for LoRA

#982 hschaeufler closed 2 months ago
5
Fix bug in upload + docs nit

#981 awni closed 2 months ago
0
I tried madlad400, but there is a problem with the output if it is float16

#980 otmb opened 2 months ago
5
Fix the cache_prompt

#979 angeloskath closed 2 months ago
0
[Feature Request] MLX_lm.cache_prompt | Save cached_prompt as plaintext in the kv-cache-file metadata

#978 mark-lord opened 2 months ago
0
Possible bug with prompt cache

#977 awni closed 2 months ago
3
is mlx-examples/llms/CONTRIBUTING.md still up-to-date?

#976 Jonathan-Dobson closed 2 months ago
1
linting / type definitions

#975 Jonathan-Dobson opened 2 months ago
0
Update LLM generation docs to use chat template

#973 awni closed 2 months ago
2
using generate() yields unusable response with llama 3.1 instruct 4bit

#972 Jonathan-Dobson closed 2 months ago
2
Fix attention layers map for SD-2-1-Base

#971 pranav4501 opened 3 months ago
2
[ Suggestion ] Add KV cache features to the Python API

#970 ibehnam closed 1 month ago
1
Make sure to import the correct "version" module when installing mlx_whisper and mlx_lm from local source code.

#969 walkoncross closed 2 months ago
7

Previous Next