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
Qwen3.8-27B
Qwen · Mike0021
Params26.9B
Context256K
Q4_K_M19.0 GB
SK
SkyReels V2 I2V 14B 540P
Skyreels · ThePun1sher
Params16.4B
Context—
Q4_K_M10.9 GB
WA
Wanabi Novelist 24B
Wanabi · kawaimasa
Params23.6B
Context128K
Q4_K_M14.3 GB
BO
Boris 75M Instruct
Boris · opencerebral
Params77M
Context1K
Q4_K_M0.1 GB
QW
Qwen3 Coder Next REAP 48B A3B
Qwen · lovedheart
Params48.9B
Context256K
Smallest22.9 GB
FL
flint MXFP4 MOE 40B A2.5B
Flint · prithivMLmods
Params39.8B
Context256K
Smallest22.2 GB
QW
Qwen3 VL 8B Thinking Heretic
Qwen · nakedlittlezombie
Params8.19B
Context256K
Q4_K_M5.0 GB
GE
Gemma-3-4B-It
Gemma · tripolskypetr
Params3.88B
Context128K
Q4_K_M2.5 GB
QW
Qwen3.5 2B heretic
Qwen · FadedRedStar
Params1.88B
Context256K
Q4_K_M1.3 GB
QW
Qwen3.5 2B_qlora_decisions De
Qwen · johannhartmann
Params1.91B
Context256K
Q4_K_M1.3 GB
DE
Cogito-V2-Preview-Deepseek-671B-Moe
DeepSeek · unsloth
Params671B
Context160K
Q4_K_M404.9 GB
QW
Qwen3 Embedding Giga Embeddings 0826 480M
Qwen · ai-babai
Params484M
Context8K
Q4_K_M0.3 GB
AQ
Aquif 3.5 Max 1205
Aquif · bartowski
Params42.4B
Context256K
Q4_K_M25.8 GB
LF
LFM2.5 VL 3B
Lfm · prithivMLmods
Params2.7B
Context125K
Q4_K_M1.7 GB
QW
Qwen3-TTS-Tokenizer-12Hz
Qwen · lookbe
Params915M
Context—
Q4_K_M0.3 GB
MU
HY-MT1.5-1.8B
Mungert · Mungert
Params1.79B
Context256K
Q4_K_M1.2 GB
QW
Qwen_Qwen3.5 9B__project Jeva 9b_1791678248
Qwen · Guilherme34
Params9.08B
Context256K
Q4_K_M5.7 GB
SP
Spark X2.6 4B
Spark · ss-lab
Params4.11B
Context1024K
Q4_K_M2.6 GB
GP
gpt-oss-20b-Derestricted
gpt-oss · Mungert
Params20.9B
Context128K
Q4_K_M16.0 GB
QW
Nanonets-OCR2-3B
Qwen · Mungert
Params3.09B
Context125K
Q4_K_M2.0 GB
NA
Nanbeige4.2 3B DSpark
Nanbeige · Anbeeld
Params848M
Context256K
Q4_K_M0.5 GB
SP
SPark X2.5 4B
Spark · lainlives
Params4.11B
Context1024K
Q4_K_M2.6 GB
DO
Docision 1.0
Docision · prithivMLmods
Params4.41B
Context256K
Q4_K_M2.7 GB
QW
Qwen3.6 27B DSV4Pro Thinking Distill
Qwen · nerkyor
Params461M
Context256K
Q4_K_M16.8 GB
AR
katanemo Arch Function Chat 1.5B
Arch · bartowski
Params1.54B
Context32K
Q4_K_M1.0 GB
QW
SoCode v1 2B
Qwen · theprint
Params1.94B
Context256K
Q4_K_M1.3 GB
SP
Spark X2.5 4B HF
Spark · stornic56
Params4.11B
Context1024K
Q4_K_M2.6 GB
QW
Qwen3.5 2B Heretic
Qwen · darrellbest
Params1.94B
Context256K
Q4_K_M1.3 GB
MI
ministral 3 3b reasoning 2512
Mistral · nile-academy
Params3.43B
Context256K
Q4_K_M2.1 GB
HE
Hemmingway 1.5 Unlocked Core
Hemmingway · brainnxdomain
Params27.3B
Context256K
Q4_K_M16.8 GB