BF16 GGUF models
Unquantized bfloat16. Original quality and the largest size. 9004 models have a BF16 file. Sizes run from 1M to 2780B parameters. The longest trained context is 10240K tokens.
QW
WhereOPD Qwen3.5 9B
Qwen · prithivMLmods
Params8.95B
Context256K
Q4_K_M5.6 GB
DE
deepseek-coder-7b-instruct-v1.5
DeepSeek · qwp4w3hyb
Params6.91B
Context4K
Q4_K_M4.2 GB
PH
Phi-4-reasoning
Phi · Mungert
Params14.7B
Context32K
Q4_K_M9.2 GB
QW
Qwythos 9B Claude Mythos 5 1M
Qwythos · Zaidusht
Params8.95B
Context1024K
Q4_K_M5.6 GB
QW
Qwen3.6 27B
Qwen · prithivMLmods
Params27.3B
Context256K
Q4_K_M16.8 GB
GL
GLM 4.6V Flash
GLM · prithivMLmods
Params9.4B
Context128K
Q4_K_M6.2 GB
JE
jevling 0.8b v1
Jevling · BricksDisplay
Params752M
Context256K
Q4_K_M0.5 GB
LL
Llama 3.1 8B UltraLong 2M Instruct
Llama · bartowski
Params8.04B
Context2096K
Q4_K_M4.9 GB
QW
RuadaptQwen3 8B Hybrid
Qwen · RefalMachine
Params8.14B
Context40K
Q4_K_M5.0 GB
FA
Falcon-H1-34B-Instruct
Falcon · unsloth
Params33.6B
Context256K
Q4_K_M20.3 GB
QW
Qwen3 Code Reasoning 4B f32
Qwen · prithivMLmods
Params4.02B
Context256K
Q4_K_M2.5 GB
GR
Granite 4.1 3b
Granite · bartowski
Params3.4B
Context128K
Q4_K_M2.2 GB
GE
Gemma 3n E4B It Absolute Heresy
Gemma · MuXodious
Params6.87B
Context32K
Q4_K_M4.2 GB
DI
DictaLM 3.0 1.7B Thinking
Dictalm · dicta-il
Params1.72B
Context61K
Q4_K_M1.1 GB
QW
Qwen3.5 Thinking AIO
Qwen · prithivMLmods
Params752M
Context256K
Smallest0.8 GB
LF
LFM2-1.2B-Extract
Lfm · Mungert
Params1.17B
Context125K
Q4_K_M0.8 GB
LF
LFM2.5-1.2B-Base
Lfm · Mungert
Params1.17B
Context125K
Q4_K_M0.7 GB
PY
Pyxidis Manim CodeGen 1.7B
Pyxidis · prithivMLmods
Params1.72B
Context40K
Q4_K_M1.1 GB
QW
Qwen3-Coder-30B-A3B-Instruct
Qwen · AskMeCodes
Params30.5B
Context256K
Q4_K_M18.6 GB
WE
WeVisOne VL 4B
Wevisone · prithivMLmods
Params4.42B
Context256K
Q4_K_M2.7 GB
MI
MioTTS 0.1B
Miotts · Aratako
Params114M
Context256K
Q4_K_M0.1 GB
PH
Phi 3 medium 128k instruct IMat
Phi · legraphista
Params14B
Context128K
Smallest3.0 GB
MA
MaralGPT Mythos 9B 2606
Maralgpt · MaralGPT
Params8.95B
Context1024K
Q4_K_M5.6 GB
FL
flux.1 lite 8B alpha
Flux · city96
Params8.16B
Context—
Smallest3.7 GB
GR
Granite 4.0 Tiny Preview BE
Granite · taronaeo
Params6.67B
Context1024K
Q4_K_M4.0 GB
GR
granite-8b-code-base-128k
Granite · Mungert
Params8.05B
Context125K
Q4_K_M4.8 GB
NE
NVIDIA Nemotron Labs 3 Puzzle 75B A9B
Nemotron · RemySkye
Params78.3B
Context1024K
Q4_K_M51.6 GB
SH
Shieldgemma 9b
Shieldgemma · legraphista
Params9.24B
Context8K
Smallest2.4 GB
LF
LFM2.5 2.6B
Lfm · nicolasembleton
Params2.7B
Context125K
Q4_K_M1.7 GB
DE
DeepSeek-R1-Distill-Llama-70B
DeepSeek · Mungert
Params70.6B
Context128K
Q4_K_M42.8 GB