Q3_K GGUF models
A GGUF quantization format. Lower numbers mean smaller files and more quality loss. 15171 models have a Q3_K file. Sizes run from 1M to 5632B parameters. The longest trained context is 102400K tokens.
MI
mlfoundations dev hp ablations mistral scheduler cosine warmup0.10 minlr1e 7
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
LL
muzammil eds Meta Llama 3.1 8B Instruct English to French v3
Llama · RichardErkhov
Params8.03B
Context128K
Q4_K_M4.9 GB
MI
dolphin-2.1-mistral-7b-snr-laser
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
LL
nkpz llama2 22b chronos alpaca experiment1
Llama · RichardErkhov
Params21.8B
Context4K
Q4_K_M13.2 GB
SM
Smollm2 135m Bootleg Instruct04
SmolLM · RichardErkhov
Params135M
Context8K
Q4_K_M0.1 GB
BL
Bloom 7b1
Bloom · RichardErkhov
Params8.1B
Context4K
Q4_K_M5.3 GB
PU
pushpendrasingh21 gpt2 medical chatbot1
Pushpendrasingh · RichardErkhov
Params163M
Context1K
Q4_K_M0.1 GB
GE
Kitsunebi v1 Gemma2 8k 9B
Gemma · RichardErkhov
Params9.24B
Context8K
Q4_K_M5.8 GB
TH
thanhtunguet gpt2 address standardizer prompted
Thanhtunguet · RichardErkhov
Params406M
Context1K
Q4_K_M0.3 GB
SP
SparseLLM ReluLLaMA 13B
Sparsellm · RichardErkhov
Params13B
Context4K
Q4_K_M7.9 GB
OP
OpenLLM Ro RoMistral 7b Instruct
Openllm · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
PH
sujithatz finbot phi 3.5 mini instruct
Phi · RichardErkhov
Params3.82B
Context4K
Q4_K_M2.4 GB
BR
BroAlanTaps GPT2 large 128 12000steps
Broalantaps · RichardErkhov
Params838M
Context1K
Q4_K_M0.5 GB
MI
Mistral 7B Instruct v0.2
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
AL
Alphacode-MALI-11B_slowtest
Alphacode · RichardErkhov
Params10.9B
Context32K
Q4_K_M6.6 GB
IN
InnovationHacksAI tkgcore
Innovationhacksai · RichardErkhov
Params8.03B
Context8K
Q4_K_M4.9 GB
CO
CodeV DS 6.7B
Codev · RichardErkhov
Params6.74B
Context16K
Q4_K_M4.1 GB
MI
CorticalStack mistral 7b openhermes 2.5 sft
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
CH
CharGen-v3-beta-rl-83-s0
Chargen · CharGen
Params23.6B
Context32K
Q4_K_M14.3 GB
GE
Gemma 2 9b
Gemma · RichardErkhov
Params9.24B
Context8K
Q4_K_M5.8 GB
QW
Qwen2.5 14B Instruct
Qwen · RichardErkhov
Params14.8B
Context32K
Q4_K_M9.0 GB
QW
Qwen2.5 1.5B Instruct
Qwen · dahara1
Params1.54B
Context32K
Q4_K_M1.0 GB
QW
Qwen2.5 14B Instruct
Qwen · dahara1
Params14.8B
Context32K
Q4_K_M9.0 GB
QW
Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive
Qwen · justfrfn
Params34.7B
Context256K
Q4_K_M21.2 GB
UN
Unbabel TowerInstruct 13B v0.1
Unbabel · RichardErkhov
Params13B
Context4K
Q4_K_M7.9 GB
LL
Meta Llama 3.1 8B
Llama · RichardErkhov
Params8.03B
Context128K
Q4_K_M4.9 GB
MI
mlfoundations dev hp ablations grid mistral bsz1024 lr2e 6 scheduler cosine warmup0.15
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
LL
LION LLaMA 3 8b Dpo v1.0
Llama · RichardErkhov
Params8.03B
Context8K
Q4_K_M4.9 GB
MI
mlfoundations dev hp ablations mistral scheduler linear warmup0.10
Mistral · RichardErkhov
Params7.24B
Context32K
Q4_K_M4.4 GB
PH
bunnycore Phi 2 OmniMatrix
Phi · RichardErkhov
Params2.78B
Context2K
Q4_K_M1.7 GB