RunsOnMy ModelsGPUs & MacsAPI pricesVRAM calculatorNew

What AI models run on the 2x RTX 4090?

The 2x RTX 4090 has 2× 24 GB of VRAM (1,008 GB/s). About 45.6 GB is usable for a model. 214 of 285 open models we track fit at 8K context.

214models fit (8K ctx)
212still fit at 32K ctx
127Blargest that fits
1008 GB/smemory bandwidth

Models that fit, largest first

ModelParamsBest quant (8K)Needs32K contextMax tok/s (est.)
Ling 3.0 flash127B (5.6B act.)Q2_K43.8 GBQ2_K499details
Ornith 1.5 122B125B (13B act.)Q2_K44.7 GBQ2_K211details
GLM 4.5 Air110B (15B act.)Q2_K44.3 GBno164details
Llama 4 Scout 17B 16E Instruct109B (18B act.)Q2_K39.1 GBQ2_K153details
Llama 4 Scout 17B 16E109B (18B act.)Q2_K40.2 GBQ2_K149details
GLM 4.5V108B (14B act.)Q2_K39.7 GBQ2_K188details
GLM 4.6V108B (14B act.)Q2_K43.0 GBno173details
Hunyuan A13B Instruct80B (13B act.)Q3_K_M37.6 GBQ3_K_M162details
Kimi Dev 72B73BQ3_K_M38.3 GBQ2_K27details
Llama 3.3 70B Instruct71BQ3_K_M35.3 GBQ2_K29details
DeepSeek R1 Distill Llama 70B71BQ3_K_M35.3 GBQ2_K29details
Llama 3.1 70B71BQ3_K_M35.1 GBQ2_K29details
Kimi Linear 48B A3B Instruct49B (3.0B act.)Q6_K38.6 GBQ6_K408details
Nous Hermes 2 Mixtral 8x7B DPO47B (13B act.)Q6_K37.2 GBQ6_K95details
falcon 40b42BQ6_K33.6 GBQ6_K29details
flint40B (1.3B act.)Q8_040.3 GBQ8_0758details
Maxwell V1 Flash37B (2.3B act.)Q8_037.6 GBQ8_0418details
Hermes 4.3 36B36BQ8_038.5 GBQ6_K26details
Seed OSS 36B Instruct36BQ8_038.5 GBQ6_K26details
Qwen3.6 35B A3B36B (3.0B act.)Q8_036.2 GBQ8_0320details
Ornith 1.5 35B A3B36B (3.0B act.)Q8_036.2 GBQ8_0319details
Qwen3.5 35B A3B36B (3.0B act.)Q8_035.4 GBQ8_0327details
Axion 1.0 35B35B (3.9B act.)Q8_035.8 GBQ8_0243details
Qwen AgentWorld 35B A3B35B (3.0B act.)Q8_035.4 GBQ8_0316details
Yi 34B 200K34BQ8_036.4 GBQ6_K28details
Yi 34B34BQ8_036.4 GBQ6_K28details
Yi 34B Chat34BQ8_036.4 GBQ6_K28details
Yi 1.5 34B Chat34BQ8_036.4 GBQ6_K28details
Nous Hermes 2 Yi 34B34BQ8_036.4 GBQ6_K28details
DeepSeek R1 Distill Qwen 32B33BQ8_035.1 GBQ6_K29details
Qwen3 32B33BQ8_035.1 GBQ6_K29details
GLM Z1 32B 041433BQ8_033.4 GBQ8_029details
GLM 4 32B 041433BQ8_033.4 GBQ8_029details
granite 4.0 h small32B (8.8B act.)Q8_032.6 GBQ8_0108details
gemma 4 31B it31BQ8_033.3 GBQ8_031details
GLM 4.7 Flash31B (4.6B act.)Q8_030.8 GBQ8_0214details
Qwen3 Coder 30B A3B Instruct31B (3.0B act.)Q8_031.7 GBQ8_0316details
North Mini Code 1.030B (3.3B act.)Q8_031.6 GBQ8_0287details
Hy Memory 30B A3B30B (3.0B act.)Q8_031.1 GBQ8_0316details
Muse Glimmer 30B30BQ8_028.5 GBQ8_034details
granite 4.2 30b29BQ8_031.6 GBQ8_032details
Qwen3.8 27B28BQ8_028.4 GBQ8_035details
Qwen3.6 27B28BQ8_028.0 GBQ8_035details
Qwen3.5 27B28BQ8_028.0 GBQ8_035details
gemma 3 27b it27BQ8_028.7 GBQ8_035details
Arete 27B27BQ8_028.1 GBQ8_035details
MiniMax H3 Image PE27BQ8_028.0 GBQ8_035details
Trinity Mini26B (2.0B act.)Q8_026.9 GBQ8_0482details
GLiDE 27B26BQ8_027.2 GBQ8_036details
diffusiongemma 26B A4B it26B (4.0B act.)Q8_026.4 GBQ8_0242details
gemma 4 26B A4B it26B (4.0B act.)Q8_026.4 GBQ8_0242details
LFM2.5 24B A2B24B (2.0B act.)Q8_024.5 GBQ8_0474details
ERNIE 4.5 21B A3B PT22BQ8_022.6 GBQ8_043details
ERNIE 4.5 21B A3B Thinking22BQ8_022.6 GBQ8_043details
gpt oss safeguard 20b22B (4.8B act.)F1613.8 GBF16328details
gpt oss 20b21B (4.2B act.)F1613.8 GBF16365details
Groot 19B A3B 26102319B (3.0B act.)F1636.4 GBF16172details
Kimi VL A3B Thinking 250616B (3.0B act.)Q8_016.8 GBQ8_0325details
Kimi VL A3B Instruct16B (3.0B act.)F1631.5 GBF16168details
Kimi VL A3B Thinking16B (3.0B act.)F1631.5 GBF16168details
deepseek moe 16b base16B (2.8B act.)F1632.8 GBF16178details
LLaDA2.1 mini16B (1.4B act.)F1631.3 GBF16352details
Moonlight 16B A3B Instruct16B (3.0B act.)F1630.7 GBF16168details
DeepSeek Coder V2 Lite Instruct16B (2.7B act.)F1630.1 GBF16189details
DeepSeek V2 Lite16B (2.7B act.)F1630.1 GBF16189details
DeepSeek V2 Lite Chat16B (2.7B act.)F1630.1 GBF16189details
Qwen2.5 Coder 14B Instruct15BF1629.7 GBF1634details
DeepSeek R1 Distill Qwen 14B15BF1629.7 GBF1634details
Qwen3 14B15BF1629.4 GBF1634details
phi 415BF1629.5 GBF1634details
gemma 3 12b it12BF1623.6 GBF1643details
Llama Guard 4 12B12BF1624.6 GBF1642details
Llama 3.2 11B Vision Instruct11BF1620.1 GBF1652details
Llama 3.2 11B Vision11BF1621.7 GBF1647details
GLM 4.6V Flash10BF1618.5 GBF1654details
GLM 4.1V 9B Thinking10BF1618.5 GBF1654details
Step3 VL 10B10BF1632.3 GBF1631details
SenseNova Loop 9B Preview9.7BF1617.8 GBF1656details
Qwen3.5 9B9.7BF1617.8 GBF1656details
Ornith 1.5 9B9.7BF1617.8 GBF1656details
Qwen3.5 9B Base9.7BF1617.8 GBF1656details
MiMo V2.6 Distill Qwen 9B9.4BF1617.8 GBF1656details
glm 4 9b9.4BF1623.2 GBF1654details
internlm3 8b instruct8.8BF1617.4 GBF1657details
granite 4.2 8b8.8BF1618.2 GBF1657details
granite 4.1 8b8.8BF1618.2 GBF1657details
Qwen3 VL 8B Instruct8.8BF1617.1 GBF1662details
Molmo2 8B8.7BF1617.1 GBF1662details
LFM2.5 8B A1B8.5B (1.0B act.)F1632.3 GBF16252details
Qwen2.5 VL 7B Instruct8.3BF1615.3 GBF1666details
UI TARS 1.5 7B8.3BF1615.3 GBF1666details
olmOCR 2 7B 10258.3BF1615.3 GBF1666details
olmOCR 7B 0225 preview8.3BF1616.6 GBF1661details
UI TARS 7B DPO8.3BF1615.3 GBF1666details
Qwen3 8B8.2BF1617.1 GBF1662details
DeepSeek R1 0528 Qwen3 8B8.2BF1617.1 GBF1662details
granite 3.3 8b instruct8.2BF1617.0 GBF1662details
granite 3.1 8b instruct8.2BF1617.0 GBF1662details
granite 3.2 8b instruct8.2BF1617.0 GBF1662details
Apertus 8B Instruct 25098.1BF1616.7 GBF1663details
Llama 3.1 8B Instruct8.0BF1616.6 GBF1663details
Llama 3.1 8B8.0BF1616.6 GBF1663details
Hermes 3 Llama 3.1 8B8.0BF1616.6 GBF1663details
DeepSeek R1 Distill Llama 8B8.0BF1631.6 GBF1631details
Meta Llama 3 8B Instruct8.0BF1616.6 GBF1663details
Meta Llama Guard 2 8B8.0BF1616.6 GBF1663details
Meta Llama 3.1 8B Instruct8.0BF1616.6 GBF1663details
Ling 3.0 tiny7.9B (1.4B act.)F1615.6 GBF16364details
MiMo 7B RL7.8BF1616.4 GBF1664details
internlm2 5 7b chat7.7BF1616.0 GBF1665details
internlm2 chat 7b7.7BF1616.0 GBF1665details
Qwen2.5 7B Instruct7.6BF1615.3 GBF1666details
Qwen2.5 Coder 7B Instruct7.6BF1615.3 GBF1666details
DeepSeek R1 Distill Qwen 7B7.6BF1615.3 GBF1666details
Falcon H1R 7B7.6BF1615.1 GBF1666details
Olmo 3 7B Instruct7.3BF1616.7 GBF1669details
Olmo 3 1025 7B7.3BF1616.7 GBF1669details
Olmo 3 7B Think7.3BF1616.7 GBF1669details
Mistral 7B v0.17.2BF1614.9 GBF1670details
Mistral 7B Instruct v0.27.2BF1614.9 GBF1670details
Mistral 7B Instruct v0.17.2BF1614.9 GBF1670details
falcon 7b7.2BF1618.4 GBF1670details
falcon 7b instruct7.2BF1618.4 GBF1670details
granite 4.0 h tiny6.9B (1.5B act.)F1613.6 GBF16344details
OLMoE 1B 7B 09246.9B (1.3B act.)F1614.4 GBF16393details
deepseek coder 7b instruct v1.56.9BF1617.2 GBF1673details
deepseek coder 6.7b instruct6.7BF1617.0 GBF1675details
deepseek coder 6.7b base6.7BF1617.0 GBF1675details
Llama 2 7b hf6.7BF1617.0 GBF1675details
Yi 6B 200K6.1BF1612.3 GBF1683details
Yi 6B6.1BF1612.3 GBF1683details
Molmo2 4B4.9BF1610.0 GBF16114details
WeVisOne VL 4B4.8BF1610.0 GBF16114details
Mage VL Cowork4.7BF1610.0 GBF16114details
Qianfan OCR4.7BF1610.1 GBF16114details
Qwen3.5 4B4.7BF169.0 GBF16120details
S5 nano4.7BF169.2 GBF16116details
LightOnOCR 3 4B4.5BF169.0 GBF16120details
SingVL S1 4B4.5BF169.0 GBF16120details
Intern Decision 4B4.5BF169.0 GBF16120details
Qwen3 VL 4B Instruct4.4BF169.3 GBF16125details
gemma 3 4b it4.3BF168.4 GBF16130details
medgemma 1.5 4b it4.3BF168.4 GBF16130details
medgemma 4b it4.3BF168.4 GBF16130details
gemma 3 4b pt4.3BF169.2 GBF16117details
Phi 3.5 vision instruct4.1BF1611.2 GBF16122details
Qwen3 4B4.0BF169.3 GBF16125details
Qwen3 4B Instruct 25074.0BF169.3 GBF16125details
granite vision 4.1 4b4.0BF167.6 GBF16148details
Phi 4 mini instruct3.8BF168.9 GBF16131details
Phi 3 mini 4k instruct3.8BF1610.6 GBF16132details
Phi 3.5 mini instruct3.8BF1610.6 GBF16132details
Phi 3 mini 128k instruct3.8BF1610.6 GBF16132details
Qwen2.5 VL 3B Instruct3.8BF166.7 GBF16163details
granite 4.2 3b3.7BF168.0 GBF16138details
granite 4.1 3b llmon preview3.7BF168.0 GBF16138details
Raon 1 nano3.5BF167.0 GBF16161details
granite 4.1 3b3.4BF167.6 GBF16148details
granite 4.0 micro3.4BF167.6 GBF16148details
DeepSeek OCR 23.4B (1.2B act.)F166.6 GBF16487details
deepseek vl2 tiny3.4B (1.2B act.)F167.4 GBF16429details
Unlimited OCR 23.3B (1.1B act.)F166.6 GBF16499details
Unlimited OCR3.3B (1.1B act.)F166.6 GBF16502details
DeepSeek OCR3.3B (1.1B act.)F166.6 GBF16502details
granite 3.1 3b a800m instruct3.3B (883M act.)F167.1 GBF16570details
Llama 3.2 3B Instruct3.2BF1613.5 GBF1678details
Llama 3.2 3B3.2BF167.5 GBF16157details
Hermes 3 Llama 3.2 3B3.2BF167.5 GBF16157details
d1 3B3.1BF1610.8 GBF1693details
LFM2.5 VL 3B3.1BF1610.8 GBF1693details
Qwen2.5 3B Instruct3.1BF166.7 GBF16163details
SmolLM3 3B3.1BF166.9 GBF16164details
SmolLM3 3B Base3.1BF166.9 GBF16164details
phi 22.8BF168.2 GBF16181details
LFM2.5 2.6B2.7BF1610.8 GBF1693details
granite 3.3 2b instruct2.5BF165.8 GBF16199details
MiniCPM5 2B2.5BF165.7 GBF16200details
WeVisOne VL 2B2.4BF165.4 GBF16247details
Qwen3.5 2B2.3BF164.5 GBF16267details
SmolVLM Instruct2.2BF165.4 GBF16278details
Qwen3 1.7B2.0BF164.8 GBF16292details
Qwen3.8 27B DSpine b162.0BF164.8 GBF16248details
d1 2B1.9BF167.2 GBF16145details
LFM3 VL 2B1.9BF167.2 GBF16145details
DeepSeek R1 Distill Qwen 1.5B1.8BF164.2 GBF16283details
LFM3 1.7B1.7BF167.2 GBF16145details
LFM3 1.7B Base1.7BF164.0 GBF16292details
SmolLM2 1.7B Instruct1.7BF165.2 GBF16294details
SmolLM2 1.7B1.7BF165.2 GBF16294details
SmolLM 1.7B1.7BF165.2 GBF16294details
granite 4.0 1b1.6BF164.3 GBF16309details
Qwen2.5 1.5B Instruct1.5BF163.8 GBF16326details
OLMo 2 0425 1B1.5BF164.4 GBF16339details
deepseek coder 1.3b instruct1.3BF164.5 GBF16374details
Llama 3.2 1B1.2BF163.2 GBF16408details
Llama 3.2 1B Instruct1.2BF165.5 GBF16203details
LFM2.5 1.2B Instruct1.2BF165.0 GBF16215details
HunyuanOCR1.1BF163.5 GBF16450details
PaddleOCR VL 1.6959MF162.5 GBF16526details
Qwen3.5 0.8B873MF162.4 GBF16665details
LightOnOCR 3 0.8B853MF162.4 GBF16665details
FastVLM 0.5B759MF162.2 GBF16664details
Qwen3 0.6B752MF162.7 GBF16841details
GOT OCR2 0716MF162.8 GBF16704details
Nemotron 3.5 Parse631MF161.9 GBF16771details
GOT OCR 2.0 hf561MF162.5 GBF16899details
SmolVLM2 500M Video Instruct507MF161.6 GBF161229details
SmolVLM 500M Instruct507MF161.6 GBF161229details
Qwen2.5 0.5B Instruct494MF161.7 GBF161020details
LFM2.5 VL 450M449MF161.9 GBF16708details
SmolLM2 360M Instruct362MF161.5 GBF161389details
SmolLM2 360M362MF161.5 GBF161389details
LFM2.5 350M354MF161.9 GBF16708details
LFM2 350M354MF161.4 GBF161422details

Just too big at 4-bit

These need up to 35% more memory than the 2x RTX 4090 has at Q4_K_M and 8K context. A smaller quant or shorter context may work.

ModelQ4_K_M needsShort by
Hunyuan A13B Instruct47.1 GB1.5 GB
Mistral 7B Instruct v0.247.7 GB2.1 GB
gpt oss safeguard 120b59.5 GB13.9 GB
gpt oss 120b59.5 GB13.9 GB
GLM 4.5V61.4 GB15.8 GB

Questions

What is the biggest model that runs on the 2x RTX 4090?

Of the models we track, the largest that fits at 8K context is Ling 3.0 flash (127B) at Q2_K, needing about 43.8 GB.

How much of the 2x RTX 4090's memory can a model use?

About 45.6 GB: we assume 95% of the 2× 24 GB is usable after driver and display overhead.

Spec source: NVIDIA specifications. Model data: Hugging Face Hub API. Speed is a bandwidth ceiling (bandwidth ÷ bytes read per token), not a benchmark. Methodology · Report an error.