21 97

momonga PRO

mmnga

mmnga

AI & ML interests

None yet

Recent Activity

updated a model 25 minutes ago

mmnga/DeepSeek-V3-0324-experts-pertok-4-gguf

published a model about 6 hours ago

mmnga/DeepSeek-V3-0324-experts-pertok-4-gguf

updated a model 4 days ago

mmnga/qwen2.5-bakeneko-32b-instruct-v2-gguf

View all activity

Organizations

mmnga's activity

New activity in mmnga/Moonlight-16B-A3B-Instruct-gguf 28 days ago

It seems completely broken

#1 opened 28 days ago by

aguspiza

New activity in mmnga/Llama-3.3-70B-Instruct-gguf about 1 month ago

Can we hire you for quantization and fine tuning?

#1 opened 3 months ago by

rafa9

New activity in mmnga/DeepSeek-V3-slice-jp64 2 months ago

Fix

#1 opened 3 months ago by

STATIKwitak

New activity in mmnga/rinna-llama-3-youko-70b-instruct-gguf 8 months ago

Would it be possible to have an 8bit gguf?

#1 opened 8 months ago by

PurityWolf

New activity in mmnga/Llama-3.1-70B-Japanese-Instruct-2407-gguf 8 months ago

Please use split ggufs instead of splitting files manually

#1 opened 8 months ago by

lmg-anon

New activity in mmnga/Mistral-Large-Instruct-2407-gguf 8 months ago

Usage in the model card seems to be ChatML format.

#1 opened 8 months ago by

yamikumods

New activity in mmnga/Phi-3-medium-128k-instruct-gguf 10 months ago

LM Studioでのエラー

#1 opened 10 months ago by

alfredplpl

New activity in stabilityai/japanese-stablelm-2-instruct-1_6b 11 months ago

Update tokenization_arcade100k.py

#1 opened 11 months ago by

mmnga

New activity in mmnga/Llama-3-70B-japanese-suzume-vector-v0.1 11 months ago

An idea

#1 opened 11 months ago by

Cran-May

New activity in mmnga/ELYZA-japanese-Llama-2-7b-fast-instruct-gguf 11 months ago

Please tell me how did you convert this FAST model into gguf file.

#1 opened 11 months ago by

wattai

New activity in Vezora/Mistral-22B-v0.1 12 months ago

Update config.json

#3 opened 12 months ago by

mmnga

New activity in mmnga/Superswallow-7b-v0.1-gguf about 1 year ago

Differences in output from the original model

#1 opened about 1 year ago by

nitky

New activity in mmnga/Mixtral-Fusion-4x7B-Instruct-v0.1 about 1 year ago

Librarian Bot: Add moe tag to model

#3 opened about 1 year ago by

librarian-bot

New activity in mmnga/Mixtral-Extraction-4x7B-Instruct-v0.1 about 1 year ago

Librarian Bot: Add moe tag to model

#1 opened about 1 year ago by

librarian-bot

New activity in mmnga/TinyMixtral-x8-Clonebase-7b about 1 year ago

Librarian Bot: Add moe tag to model

#1 opened about 1 year ago by

librarian-bot

New activity in mmnga/Mixtral-Fusion-4x7B-Instruct-v0.1 over 1 year ago

Maybe a slerp or some other merge method will preserve the component experts better?

#2 opened over 1 year ago by

BlueNipples

Responses somewhat related to the prompt but still gibberish

#1 opened over 1 year ago by

JeroenAdam

New activity in mmnga/cyberagent-calm2-7b-chat-GPTQ-calib-ja-1k over 1 year ago

Tritonのサポート切れによるColab A100への移行

#2 opened over 1 year ago by

alfredplpl

bfloat16でなくfloat16による量子化

#1 opened over 1 year ago by

alfredplpl

New activity in elyza/ELYZA-japanese-Llama-2-7b-fast-instruct over 1 year ago

Missing tokenizer.model

#1 opened over 1 year ago by

mmnga