Finding 1

finding1's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in FakeRockert543/gemma-4-31b-it-MLX-8bit about 2 hours ago

"At least one mel filter has all zero values"

#1 opened about 2 hours ago by

finding1

New activity in mlx-community/GLM-5.2-DQ4plus-q8 14 days ago

ValueError: Missing 285 parameters

7

#1 opened 15 days ago by

finding1

New activity in inferencerlabs/DeepSeek-V3.2-MLX-Q4.8 15 days ago

ValueError: Expected shape (128, 512, 16) but received shape (128, 512, 24) for parameter model.layers.0.self_attn.embed_q.weight

1

#2 opened 15 days ago by

finding1

New activity in froggeric/Mistral-Medium-3.5-128B-MLX-4bit 2 months ago

Possibly messsed up conversion

3

#1 opened 2 months ago by

finding1

New activity in mlx-community/Qwen3.5-397B-A17B-8bit 5 months ago

Was this created with the fixes in mlx-lm version 0.30.8?

1

#1 opened 5 months ago by

finding1

New activity in meituan-longcat/LongCat-Flash-Chat 10 months ago

What sampler settings were used to achieve the reported benchmark scores?

👀 1

2

#8 opened 10 months ago by

finding1

New activity in mlx-community/ERNIE-4.5-300B-A47B-PT-4bit 11 months ago

Q8 version please

2

#2 opened 12 months ago by

acediac

New activity in unsloth/Llama-4-Scout-17B-16E-Instruct-GGUF about 1 year ago

Llama 4 GGUFs now with multimodal (image) capabilities.

9

#5 opened about 1 year ago by

shimmyshimmer

New activity in mradermacher/DeepSeek-V3-Base-GGUF about 1 year ago

Q4_K_M missing blk.8.ffn_up_exps.weight

1

#1 opened about 1 year ago by

finding1

Q4_K_M missing blk.8.ffn_up_exps.weight

1

#1 opened about 1 year ago by

finding1