mobicham's picture

mobicham

mobicham

·

https://www.mobiuslabs.com/

mobicham

AI & ML interests

Model pruning, quantization, computer vision, LLMs

Organizations

New activity in QuantTrio/Qwen3-VL-30B-A3B-Instruct-AWQ 2 months ago

Doesn't work with vllm

#1 opened 3 months ago by

New activity in dropbox-dash/Llama-3.1-8B-Instruct_gemlite-ao_a16w4_gs_128_pack_32bit 5 months ago

🙋🏻‍♂️ gemlite hqq quant version "unlearns" tool calling (?)

#2 opened 5 months ago by

New activity in dropbox-dash/Qwen2.5-VL-3B-Instruct_4bitgs64_hqq_hf 8 months ago

RuntimeError: b_zeros dim 1 = 2560 is not size_n / pack_factor = 320

#1 opened 8 months ago by

New activity in microsoft/bitnet-b1.58-2B-4T-bf16 8 months ago

Static cache support

#1 opened 8 months ago by

New activity in mobiuslabsgmbh/Llama-2-70b-hf-2bit_g16_s128-HQQ 9 months ago

no run

#1 opened 10 months ago by

New activity in mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.1 11 months ago

Open source eval metrics and library?

#1 opened 11 months ago by

New activity in mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1 11 months ago

No model file found in repo 'mobiuslabsgmbh/DeepSeek-R1-ReDistill-Llama3-8B-v1.1'

#1 opened 11 months ago by

New activity in mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-7B-v1.1 11 months ago

It's bad, sorry.

#2 opened 11 months ago by

Will there be a 32b and 70b too?

#1 opened 11 months ago by

AlgorithmicKing

New activity in mobiuslabsgmbh/DeepSeek-R1-ReDistill-Qwen-1.5B-v1.0 11 months ago

Open Source Distillation Approach?

#1 opened 11 months ago by

New activity in mobiuslabsgmbh/Mixtral-8x7B-Instruct-v0.1-hf-attn-4bit-moe-2bitgs8-metaoffload-HQQ 11 months ago

not run

#2 opened 11 months ago by

New activity in dropbox-dash/Llama-3-8b-instruct_2bitgs64_hqq 11 months ago

erorr

#2 opened 11 months ago by

the coder from the model card has errors when executing on google colab

#1 opened over 1 year ago by

New activity in dropbox-dash/Llama-3.1-8b-instruct_4bitgs64_hqq 11 months ago

not run

#1 opened 11 months ago by

New activity in mobiuslabsgmbh/Llama-2-7b-chat-hf-4bit_g64-HQQ 11 months ago

not run

#1 opened about 1 year ago by

New activity in dropbox-dash/lama-3.1-70b-instruct_4bitgs64_hqq over 1 year ago

CPU support

#2 opened over 1 year ago by

Decensored version?

#1 opened over 1 year ago by

KnutJaegersberg

New activity in dropbox-dash/Llama-3.1-8b-instruct_4bitgs64_hqq_calib over 1 year ago

Oobabooga?

#1 opened over 1 year ago by

New activity in mobiuslabsgmbh/Llama-2-7b-chat-hf_1bitgs8_hqq over 1 year ago

QUANTIZED VERSION GGUF

#5 opened over 1 year ago by

New activity in open-llm-leaderboard/open_llm_leaderboard over 1 year ago

GSM8K (5-shot) performance is quite different compared to running lm_eval locally

#755 opened over 1 year ago by