F16 GGUF models

Unquantized 16-bit floats. Original quality and the largest size. 53239 models have a F16 file. Sizes run from 0M to 1030B parameters. The longest trained context is 10742K tokens.

Most downloadedNewestLargestLongest contextSmallest
OR
Ornith1.5 9B 3MPER0RR Obliterated
Ornith · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
Vision
FI
Fireball 3.1 8B ORPO
Fireball · bartowski
Params8B
Context128K
Q4_K_M4.9 GB
GGUF
GE
Gemma4_E2B_Abliterated_Opus_Distilled
Gemma · mradermacher
Params4.63B
Context128K
Q4_K_M3.4 GB
ReasoningVision
QW
Qwen3.5 Hasutsubomi 9B
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningRoleplayTool use
MI
MiniCPM V 4.7 1B
MiniCPM · mradermacher
Params1.01B
Context256K
Q4_K_M0.7 GB
Vision
HU
Hubble 4B V1C
Hubble · bartowski
Params4.51B
Context128K
Q4_K_M2.8 GB
GGUF
LL
Llama3Ternary 2b
Llama · CMSManhattan
Params2.1B
Context8K
Q4_K_M1.3 GB
Tool use
QW
Qwen3.5 9B Deckard Claude DIMOE Uncensored Heretic Thinking
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningRoleplayVision
QW
Qwen2.5 VL 32B Instruct
Qwen · second-state
Params32.8B
Context125K
Q4_K_M19.9 GB
Vision
CA
CataLlama-v0.2-Instruct-SFT
Catallama · mradermacher
Params8.03B
Context8K
Q4_K_M4.9 GB
GGUF
GE
Gemma 4 E2B It Dspark
Gemma · rasyosef
Params262M
Context128K
Smallest0.3 GB
ReasoningTool use
GE
Gemma 4 E4B Luchador Rudo
Gemma · mradermacher
Params7.52B
Context128K
Q4_K_M5.3 GB
ReasoningRoleplayTool use
NE
Daredevil-8B-abliterated-dpomix
Neuraldaredevil · mradermacher
Params8B
Context8K
Q4_K_M4.9 GB
GGUF
MO
moss tts local v1.5
Moss · cstr
Params2.12B
Context—
Smallest2.1 GB
GGUF
HO
Holo 3.1 0.8B
Holo · mradermacher
Params1.01B
Context256K
Q4_K_M0.7 GB
ReasoningTool useVision
OP
OpenChat 3.5 7B SOLAR v2.0
Openchat · mradermacher
Params7.24B
Context8K
Q4_K_M4.4 GB
GGUF
LL
Llama 3.1 EIRAI 8B Prob
Llama · mradermacher
Params8B
Context8K
Q4_K_M4.9 GB
GGUF
EZ
Ezjev 4b S2
Ezjev · mradermacher
Params4.21B
Context256K
Q4_K_M2.7 GB
Vision
HY
HybridAgent 8B Agentic
Hybridagent · mradermacher
Params8.19B
Context32K
Q4_K_M5.0 GB
ReasoningTool use
LL
llama-joycaption-beta-one-hf-llava
Llama · Mungert
Params8.03B
Context128K
Q4_K_M5.0 GB
Tool useVision
SA
Sarvam 1
Sarvam · bartowski
Params2.53B
Context8K
Q4_K_M1.5 GB
Tool use
GE
Gemma 4 E4B It Uncensored MAX
Gemma · prithivMLmods
Params7.5B
Context128K
Smallest8.0 GB
ReasoningVision
QW
Qwen3.5 9b Sushi Coder
Qwen · mradermacher
Params8.95B
Context256K
Q4_K_M5.6 GB
CodingReasoningTool use
LF
LFM2.5 1.2B Thinking heretic
Lfm · saidutta69
Params1.17B
Context125K
Q4_K_M0.7 GB
ReasoningTool use
LL
Replete LLM V2 Llama 3.1 8b
Llama · bartowski
Params8B
Context128K
Q4_K_M4.9 GB
CodingTool use
HA
Hare1.0 Beta
Hare · mradermacher
Params1.19B
Context32K
Q4_K_M0.7 GB
GGUF
QW
Qwen2-0.5B
Qwen · mradermacher
Params494M
Context128K
Q4_K_M0.4 GB
GGUF
JA
AI21-Jamba-Reasoning-3B
Jamba · Mungert
Params3.2B
Context256K
Q4_K_M2.0 GB
ReasoningTool use
QW
Qwen3 1.7B
Qwen · second-state
Params2.03B
Context40K
Q4_K_M1.3 GB
ReasoningTool use
TR
Trinas Nectar 8B model stock
Trinas · mradermacher
Params8B
Context8K
Q4_K_M4.9 GB
GGUF

Other quants

Keep browsing