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.
OR
Ornith-1.5-9B
Ornith · Mungert
Params9.2B
Context256K
Best fitQ5_0 8.0 GB
AE
AesCode 32B
Aescode · mradermacher
Params32.8B
Context256K
Q4_K_M19.8 GB
LE
LeetCodeWizard 7B V1.1 i1
Leetcodewizard · mradermacher
Params6.74B
Context16K
Best fitIQ3_M 7.9 GB
LF
LFM3 1.7B Heretic Abliterated i1
Lfm · Vortecks
Params1.73B
Context256K
Best fitQ8_0 4.0 GB
GE
gemma 3n E4B it text
Gemma · lmstudio-community
Params6.9B
Context32K
Best fitQ6_K 6.8 GB
QW
Qwen Image 2.1 Anime Fusal
Qwen · LocalAI-io
Params7.12B
Context256K
Best fitQ5_0 6.8 GB
MI
Mistral Helcyon Mercury 12b v3.2 Absolute Heresy
Mistral · mradermacher
Params12.2B
Context128K
Best fitQ3_K_M 7.9 GB
YI
01 ai Yi 6B 200K
Yi · RichardErkhov
Params6.06B
Context195K
Best fitQ6_K 6.0 GB
LL
Llama Guard 3 8B
Llama · mradermacher
Params8B
Context128K
Best fitQ5_K_M 7.3 GB
GR
Ibm Granite_Granite 4.0 H Tiny
Granite · lmstudio-community
Params6.94B
Context1024K
Best fitQ4_K_M 6.7 GB
PH
Athene Phi 3.5 mini instruct orpo i1
Phi · mradermacher
Params3.82B
Context128K
Best fitQ6_K 6.9 GB
NE
NeoHorse 1 9B Math SFT
Neohorse · mradermacher
Params8.95B
Context256K
Best fitQ5_K_M 8.0 GB
LL
codellama_codellama-7b-python-hf
Llama · TheBloke
Params6.74B
Context16K
Best fitQ3_K_S 7.7 GB
AR
AREX 2 i1
Arex · mradermacher
Params27B
Context256K
Q4_K_M16.5 GB
ZE
Zephyr 7B Beta Harmless
Zephyr · mradermacher
Params7.24B
Context32K
Best fitQ6_K 7.5 GB
GL
GLM 5.3 Flash DFlash2
GLM · Anbeeld
Params1.17B
Context1024K
Best fitBF16 3.0 GB
LL
Llama 3.2 3B Instruct
Llama · legraphista
Params3.21B
Context128K
Best fitBF16 7.9 GB
OR
Ornith 1.5 35B A3B Uncensored
Ornith · mradermacher
Params35.5B
Context256K
Q4_K_M21.7 GB
JE
Jev-Style-2B-Decision-v3
Jev · chaoliangUNSW
Params1.88B
Context256K
Best fitF16 4.7 GB
MI
Ministral 3 3B Reasoning 2512
Mistral · mistralai
Params3.43B
Context256K
Best fitQ8_0 5.0 GB
LL
LogDoctor Llama 3.1 8B
Llama · mradermacher
Params8B
Context128K
Best fitQ5_K_M 7.3 GB
QW
qwen3.5 9B super coder
Qwen · jica98
Params9.2B
Context256K
Best fitQ4_0 7.1 GB
DI
discoresearch_discolm_german_7b_v1
Discolm · TheBloke
Params7.24B
Context32K
Best fitQ6_K 7.5 GB
OR
Ornith 1.5 122B DFlash2
Ornith · kaivoss
Params930M
Context256K
Best fitF32 4.4 GB
QW
Qwen3Vl 4b Thinking
Qwen · Qwen
Params4B
Context256K
Best fitQ8_0 6.0 GB
IN
InternVL3_5 14B
Internvl · bartowski
Params14.8B
Context40K
Best fitIQ3_XXS 7.8 GB
SA
SauerkrautLM 7b LaserChat
Sauerkrautlm · mradermacher
Params7.24B
Context8K
Best fitQ6_K 7.5 GB
CY
Cydonia 24B v4.3 Heretic v4
Cydonia · mradermacher
Params23.6B
Context128K
Best fitIQ1_M 7.6 GB
CO
Cosmos-Reason1-7B
Cosmos · unsloth
Params7.62B
Context125K
Best fitQ6_K_XL 7.9 GB
QW
Qwen3 Vl 4b Heretic
Qwen · DreamFast
Params4B
Context256K
Best fitQ8_0 6.0 GB