GGUF models by sizzlebop

140 GGUF models quantized by sizzlebop on Hugging Face. Sizes run from 27M to 32.8B parameters. The longest trained context is 1024K tokens.

Most downloadedNewestLargestLongest contextSmallest
SP
Spark X2.5 4B
Spark · sizzlebop
Params4.11B
Context1024K
Q4_K_M2.6 GB
GGUF
QW
OxCoder 9B
Qwen · sizzlebop
Params8.95B
Context256K
Q4_K_M5.6 GB
CodingReasoningTool use
QW
Minecraft Agent Qwen3.5 9B Stage3
Qwen · sizzlebop
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool useVision
MI
Minimind 3
Minimind · sizzlebop
Params64M
Context32K
Q4_K_M0.0 GB
ReasoningTool use
SP
Spark X2.5 1.7B
Spark · sizzlebop
Params1.71B
Context1024K
Q4_K_M1.1 GB
ReasoningTool use
PR
PrimeMind 9B Bf16
Primemind · sizzlebop
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool useVision
QW
CodeThink V4 Qwen3 4B Mix
Qwen · sizzlebop
Params4.41B
Context32K
Q4_K_M2.7 GB
CodingReasoningTool use
EV
EvoLM 8B Rubric
Evolm · sizzlebop
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
QW
CodeThink V4 Qwen3 8B
Qwen · sizzlebop
Params8.19B
Context32K
Q4_K_M5.0 GB
CodingReasoningTool use
HA
Hayabusa 9b
Hayabusa · sizzlebop
Params8.95B
Context256K
Q4_K_M5.6 GB
ReasoningTool use
LL
Llama 3.2 3B TechWriter Instruct
Llama · sizzlebop
Params3.21B
Context128K
Q4_K_M2.0 GB
Tool use
OR
ORP 0.8B
Orp · sizzlebop
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningRoleplayTool use
HY
HybridIntelligence 0.5B
Hybridintelligence · sizzlebop
Params521M
Context16K
Q4_K_M0.3 GB
GGUF
NE
Nebula
Nebula · sizzlebop
Params321M
Context4K
Q4_K_M0.2 GB
Reasoning
TI
Pip v1.0 TinyReasoner
Tinyllama · sizzlebop
Params1.1B
Context8K
Q4_K_M0.7 GB
Reasoning
GL
Glimmer 1 Base
Glimmer · sizzlebop
Params—
Context512
Q4_K_M0.0 GB
GGUF
TI
tinybrainbot 350mV3 instruct
Tinybrainbot · sizzlebop
Params348M
Context2K
Q4_K_M0.3 GB
GGUF
TI
tinybrainbot 350mV3 base
Tinybrainbot · sizzlebop
Params348M
Context2K
Q4_K_M0.3 GB
GGUF
RE
H 05b Replay
Replay · sizzlebop
Params521M
Context16K
Q4_K_M0.3 GB
GGUF
ON
Onca 3.0 12b
Onca · sizzlebop
Params11.9B
Context256K
Q4_K_M7.4 GB
GGUF
FL
Flame 27m Instruct
Flame · sizzlebop
Params27M
Context2K
Q4_K_M0.0 GB
ReasoningTool use
RI
RimTalk Mini v1
Rimdialogue · sizzlebop
Params3.21B
Context128K
Smallest1.9 GB
GGUF
RI
RimTalk Mini v1
Rimdialogue · sizzlebop
Params3.21B
Context128K
Smallest3.4 GB
GGUF
CR
Crystal Think V2
Crystal · sizzlebop
Params4.02B
Context32K
Q4_K_M2.5 GB
Reasoning
NA
Nanonets OCR S
Nanonets · sizzlebop
Params3.09B
Context125K
Smallest3.3 GB
Vision
QW
Qwen3.5 9B Fable 5 v1
Qwen · sizzlebop
Params9.2B
Context256K
Smallest6.6 GB
CodingReasoningTool use
CU
CursedMatrix 8B v9 IQ4 NL
Cursedmatrix · sizzlebop
Params8.03B
Context8K
Smallest4.7 GB
Roleplay
MI
Mini Coder 1.7b
Mini · sizzlebop
Params1.72B
Context40K
Smallest1.1 GB
Coding
GR
Granite 4.0 H Tiny
Granite · sizzlebop
Params6.94B
Context1024K
Smallest4.0 GB
GGUF
KU
Kuwutu 7B
Kuwutu · sizzlebop
Params7.62B
Context32K
Smallest4.5 GB
CodingReasoningRoleplay

About sizzlebop on Hugging Face

sizzlebop has published 140 GGUF models that ggufmodels.com tracks, across families such as Qwen, Granite and Crystal. The most downloaded are Spark X2.5 4B, OxCoder 9B and Minecraft Agent Qwen3.5 9B Stage3.

The quants sizzlebop publishes most often are Q8_0 (91), Q4_K_M (43), Q5_K_M (27) and Q6_K (21). Their Q4_K_M files range from 0.0 to 19.9 GB. Parameter counts run from 27M to 32.8B, and the longest trained context window is 1024K tokens.

Frequently asked questions

What GGUF models has sizzlebop published?

140 models are listed here, mainly from the Qwen, Granite and Crystal families. The most downloaded is Spark X2.5 4B.

Which quants does sizzlebop publish?

The most common are Q8_0 (91 models), Q4_K_M (43 models), Q5_K_M (27 models) and Q6_K (21 models).

Keep browsing