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.

Most downloadedNewestLargestLongest contextSmallest
NE
OpenMath Nemotron 14B
Nemotron · bartowski
Params14.8B
Context128K
Q4_K_M9.0 GB
Reasoning
QW
GameMaster v1 2B
Qwen · theprint
Params1.94B
Context256K
Q4_K_M1.3 GB
ReasoningTool useVision
VI
Violet Twilight v0.2 GGUF IQ Imatrix
Violet · Lewdiculous
Params12.2B
Context1000K
Q4_K_M7.5 GB
Roleplay
US
UserRepGen 4B
Userrepgen · prithivMLmods
Params4.21B
Context256K
Q4_K_M2.7 GB
ReasoningTool use
GE
Gemma 4 26B A4B It Ultra Uncensored Heretic
Gemma · Arue-2026
Params25.2B
Context256K
Q4_K_M16.8 GB
ReasoningTool useVision
QW
Qwen3.6 27B Abliterated Heretic Uncensored
Qwen · marafx2025
Params26.9B
Context256K
Q4_K_M16.5 GB
ReasoningTool useVision
GR
granite-4.0-tiny-preview
Granite · Mungert
Params6.67B
Context1024K
Q4_K_M4.2 GB
ReasoningTool use
AP
Walnoot 8B Instruct
Apertus · WAINUT
Params8.05B
Context64K
Smallest16.1 GB
GGUF
LL
Llama-3.1-8B-Think
Llama · bartowski
Params8.03B
Context128K
Q4_K_M4.9 GB
ReasoningTool use
SE
SERA 8B GA
Sera · bartowski
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
BY
ByteCraft
Bytecraft · bartowski
Params8.39B
Context32K
Q4_K_M5.2 GB
Tool use
QW
Qwen3.5 4B DS v4 Flash v1.0
Qwen · prithivMLmods
Params4.21B
Context256K
Q4_K_M2.7 GB
ReasoningTool useVision
QW
Qwen2.5 Coder 0.5B
Qwen · Volko76
Params494M
Context32K
Q4_K_M0.4 GB
CodingTool use
OL
Olmo Marvin Full Merged
OLMo · bartowski
Params32.2B
Context64K
Q4_K_M19.5 GB
Tool use
QW
Qwen3.8 Flash Next
Qwen · Delgon
Params51.2B
Context—
Smallest28.8 GB
GGUF
GE
gemma 4 E4B it GMCE
Gemma · emperorofrome
Params7.52B
Context128K
Q4_K_M5.3 GB
ReasoningTool use
GR
Granite 4.0 H 1b Base
Granite · ibm-granite
Params1.46B
Context1024K
Q4_K_M0.9 GB
GGUF
QW
Qwen3.5 2B heretic imatrix
Qwen · FadedRedStar
Params331M
Context256K
Q4_K_M1.3 GB
CodingVision
AM
amos 1 medium
Amos · prithivMLmods
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool useVision
GE
Gemma 4 26B A4B It
Gemma · axteam
Params25.2B
Context256K
Q4_K_M16.8 GB
ReasoningTool useVision
QW
QwenDGAF V3 0.6B
Qwen · hecan3d
Params596M
Context40K
Q4_K_M0.4 GB
ReasoningTool use
GE
Gemma 4 E2B It Qat Q4_0 Unquantized
Gemma · WhiskyAKM
Params4.63B
Context128K
Q4_K_M3.4 GB
ReasoningTool useVision
GR
granite-3b-code-instruct-2k
Granite · Mungert
Params3.48B
Context2K
Q4_K_M2.1 GB
CodingTool use
SE
Seed-Coder-8B-Reasoning
Seed · Mungert
Params8.25B
Context64K
Q4_K_M5.2 GB
CodingReasoning
QW
Qwen Image Bench
Qwen · DawnW0lf
Params26.9B
Context256K
Q4_K_M16.5 GB
ReasoningTool useVision
QW
Qwen3 4B Thinking 2507 GPT 5.1 Codex Max Distill
Qwen · TeichAI
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use
RW
RWKV7-Goose-World3-2.9B-HF
Rwkv · Mungert
Params2.95B
Context1024K
Q4_K_M1.9 GB
GGUF
QW
Deductive Reasoning Qwen 14B
Qwen · bartowski
Params14.8B
Context32K
Q4_K_M9.0 GB
ReasoningTool use
DE
Decision 2.0 Lux 9B
Decision · vllm-sr
Params7.94B
Context16K
Q4_K_M5.1 GB
ReasoningTool use
DE
Qwen3 4B Thinking 2507 DeepSeek v3.2 Speciale Math Distill
DeepSeek · TeichAI
Params4.02B
Context256K
Q4_K_M2.5 GB
ReasoningTool use

Other quants

Keep browsing