GGUF models that fit in 8 GB
151917 GGUF language models have at least one quant whose estimated memory use is 8 GB or less at 8K context. Each card shows the largest quant that fits. Sizes run from 1M to 20260731B parameters. The longest trained context is 4194304K tokens.
QW
Qwen2.5 7B Instruct Abliterated
Qwen · mradermacher
Params7.62B
Context32K
Best fitQ6_K 7.2 GB
GE
Huihui gemma 4 26B A4B it qat q4 0 unquantized abliterated
Gemma · huihui-ai
Params25.2B
Context128K
Best fitBF16 2.4 GB
TE
TerminalHorizon 122B A10B
Terminalhorizon · mradermacher
Params122B
Context256K
Q4_K_M74.2 GB
LF
LFM2.5-VL-3B
Lfm · LiquidAI
Params2.7B
Context32K
Best fitBF16 7.9 GB
PR
Fish Speech S2 Pro Dual AR
Pro · rodrigomt
Params4.95B
Context32K
Best fitQ8_0 6.9 GB
LL
tiny llamas
Llama · ggml-org
Params24M
Context—
Smallest0.0 GB
HO
K2 Horizon 7B
Horizon · IFM
Params9B
Context512K
Best fitQ5_0 8.0 GB
QW
Qwen2.5 7B Instruct abliterated v2
Qwen · QuantFactory
Params7.6B
Context32K
Best fitQ6_K 7.2 GB
LF
LFM2.5 VL 1.6B
Lfm · LiquidAI
Params1.17B
Context125K
Best fitBF16 3.9 GB
QW
Qwen3.5 Text 9B NSFW RP RolePlay
Qwen · mradermacher
Params8.95B
Context256K
Best fitQ5_K_M 8.0 GB
LL
Meta-Llama-3-8B-Instruct
Llama · bartowski
Params8B
Context8K
Best fitQ5_K_M 7.3 GB
MI
Mistral Nemo Instruct 2407
Mistral · bartowski
Params12.2B
Context1000K
Best fitQ3_K_M 7.9 GB
QW
Qwen3 8B Abliterated
Qwen · bartowski
Params8.19B
Context40K
Best fitQ5_K_L 7.9 GB
DE
DeepSeek V4.1 Flash 1017 abliterated
DeepSeek · orcarouter
Params748B
Context1024K
Smallest1.6 GB
NE
NVIDIA Nemotron 3.5 Super VL 120B A12B
Nemotron · unsloth
Params1.63B
Context1024K
Smallest1.3 GB
SM
SmolLM3 3B
SmolLM · ggml-org
Params3.08B
Context64K
Best fitF16 7.3 GB
QW
LFM2.5 8B A1B
Qwen · DavidAU
Params8.47B
Context125K
Best fitIQ4_XS 7.0 GB
MX
mxbai embed xsmall v1
Mxbai · mixedbread-ai
Params24M
Context4K
Best fitF32 0.7 GB
QW
Qwen3.5 0.8B
Qwen · bartowski
Params773M
Context256K
Best fitBF16 2.5 GB
ST
stock report2
Stock · uiyong
Params—
Context—
Best fitQ8_0 0.5 GB
QW
Qwen3.5 9B Ultra Uncensored Heretic
Qwen · llmfan46
Params8.95B
Context256K
Best fitQ5_K_M 8.0 GB
GE
Huihui gemma 4 31B it qat q4 0 unquantized abliterated
Gemma · huihui-ai
Params30.7B
Context128K
Best fitBF16 3.6 GB
KA
Karna 49M SFT Model
Karna · VarunHujan
Params72M
Context2K
Best fitQ8_0 0.7 GB
SA
Sage-A1.5
Sage · na0x2c6
Params35.5B
Context256K
Q4_K_M21.7 GB
EM
Embeddinggemma 300m
Embeddinggemma · ente-ai
Params308M
Context2K
Best fitQ8_0 1.0 GB
GE
gemma-2-9b-it
Gemma · bartowski
Params9.24B
Context8K
Best fitIQ3_M 7.8 GB
OR
Ornith 1.5 35B A3B
Ornith · AtomicChat
Params34.7B
Context256K
Best fitQ8_0 3.2 GB
LF
LFM3 1.7B Uncensored
Lfm · AtomicChat
Params1.73B
Context256K
Best fitQ6_K 3.7 GB
MA
Magidonia 24B v4.3
Magidonia · bartowski
Params23.6B
Context128K
Q4_K_M14.3 GB
SA
Sao10K L3 8B Stheno v3.1
Sao · RichardErkhov
Params8B
Context—
Best fitIQ4_NL 5.8 GB