RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?OllamaStart hereAPI pricesVRAM calculatorNew

What AI models run on the RTX 4060 Ti 16GB?

The RTX 4060 Ti 16GB has 16 GB of VRAM (288 GB/s). About 15.2 GB is usable for a model. 200 of 298 open models we track fit at 8K context.

200models fit (8K ctx)
189still fit at 32K ctx
37Blargest that fits
288 GB/smemory bandwidth

This card has 16 GB of VRAM. See every model that fits in 16 GB VRAM. Other 16 GB hardware: RTX 4070 Ti Super, RTX 4080, RTX 4080 Super, RTX 5060 Ti 16GB, RTX 5070 Ti, RTX 5080, RX 7800 XT, RX 7900 GRE.

Compare for local LLMs: RTX 4060 Ti 16GB vs RTX 4060 Ti 8GBRTX 4060 Ti 16GB vs RTX 4070RTX 4060 Ti 16GB vs RTX 3060 12GBRTX 4060 Ti 16GB vs RTX 5060 Ti 16GBRTX 4060 Ti 16GB vs RX 7800 XTRTX 4060 Ti 16GB vs Arc A770 16GBRTX 4060 Ti 16GB vs Mac M4 Pro 24GB all comparisons

Using a specific tool? Ollama models that run on the RTX 4060 Ti 16GB · LM Studio · llama.cpp offload

Models that fit, largest first

ModelParamsBest quant (8K)Needs32K contextMax tok/s (est.)Details
Maxwell V1 Flash37B (2.3B act.)Q2_K14.0 GBQ2_K338details
Qwen3.6 35B A3B36B (3.0B act.)Q2_K13.6 GBQ2_K256details
Ornith 1.5 35B A3B36B (3.0B act.)Q2_K13.2 GBQ2_K264details
Qwen3.5 35B A3B36B (3.0B act.)Q2_K13.6 GBQ2_K256details
Sage A1.536B (4.7B act.)Q2_K13.4 GBQ2_K165details
Qwen AgentWorld 35B A3B35B (3.0B act.)Q2_K13.1 GBQ2_K256details
Yi 1.5 34B Chat34BQ2_K14.3 GBno22details
Nous Hermes 2 Yi 34B34BQ2_K14.8 GBno22details
DeepSeek R1 Distill Qwen 32B33BQ2_K14.2 GBno23details
Qwen3 32B33BQ2_K14.1 GBno23details
GLM Z1 32B 041433BQ2_K12.5 GBQ2_K24details
GLM 4 32B 041433BQ2_K12.6 GBQ2_K23details
Olmo 3 32B Think32BQ2_K13.0 GBQ2_K24details
Granite 4.0 H Small32B (8.8B act.)Q2_K11.7 GBQ2_K89details
Gemma 4 31B IT31BQ2_K14.7 GBno23details
Gemma 4 31B31BQ2_K13.9 GBno25details
GLM 4.7 Flash31B (4.6B act.)Q2_K11.4 GBQ2_K177details
Qwen3 Coder 30B A3B Instruct31B (3.0B act.)Q2_K12.1 GBQ2_K256details
North Mini Code 1.030B (3.3B act.)Q2_K11.7 GBQ2_K240details
Hy Memory 30B A3B30B (3.0B act.)Q2_K11.9 GBQ2_K256details
Muse Glimmer 30B30BQ2_K11.4 GBQ2_K26details
Granite 4.2 30B29BQ2_K12.7 GBno27details
Raon 1 Mini Preview29B (4.3B act.)Q2_K11.2 GBQ2_K178details
Qwen3.8 27B28BQ2_K11.5 GBQ2_K27details
Qwen3.6 27B28BQ2_K12.6 GBQ2_K24details
Qwen3.5 27B28BQ2_K12.6 GBQ2_K24details
Gemma 3 27B IT27BQ2_K11.7 GBQ2_K27details
Arete 27B27BQ3_K_M13.6 GBQ2_K22details
MiniMax H3 Image PE27BQ2_K11.3 GBQ2_K27details
Trinity Mini26B (2.0B act.)Q3_K_M12.3 GBQ3_K_M316details
GLiDE 27B26BQ3_K_M13.2 GBQ2_K23details
DiffusionGemma 26B A4B IT26B (4.0B act.)Q3_K_M13.2 GBQ2_K147details
Gemma 4 26B A4B IT26B (4.0B act.)Q3_K_M13.5 GBQ2_K143details
LFM2.5 24B A2B24B (2.0B act.)Q3_K_M11.7 GBQ3_K_M295details
ERNIE 4.5 21B A3B PT22BQ4_K_M13.6 GBQ3_K_M21details
ERNIE 4.5 21B A3B Thinking22BQ4_K_M13.6 GBQ3_K_M21details
gpt-oss Safeguard 20B22B (4.8B act.)Q8_012.3 GBQ8_0107details
gpt-oss 20B21B (4.2B act.)Q8_012.3 GBQ8_0119details
Kimi VL A3B Thinking 250616B (3.0B act.)Q6_K13.5 GBQ5_K_M117details
Kimi VL A3B Thinking16B (3.0B act.)Q6_K13.5 GBQ5_K_M117details
Kimi VL A3B Instruct16B (3.0B act.)Q5_K_M12.2 GBQ5_K_M131details
DeepSeek MoE 16B Base16B (2.8B act.)Q4_K_M12.5 GBQ2_K154details
LLaDA2.1 Mini16B (1.4B act.)Q6_K13.4 GBQ5_K_M245details
Moonlight 16B A3B Instruct16B (3.0B act.)Q5_K_M12.2 GBQ5_K_M127details
DeepSeek Coder V2 Lite Instruct16B (2.7B act.)Q5_K_M11.9 GBQ5_K_M143details
DeepSeek V2 Lite16B (2.7B act.)Q6_K12.8 GBQ6_K132details
DeepSeek V2 Lite Chat16B (2.7B act.)Q5_K_M11.9 GBQ5_K_M143details
Qwen2.5 Coder 14B Instruct15BQ6_K13.5 GBQ3_K_M24details
DeepSeek R1 Distill Qwen 14B15BQ6_K13.5 GBQ3_K_M24details
Qwen3 14B15BQ6_K13.2 GBQ3_K_M24details
Phi 415BQ6_K13.4 GBQ2_K24details
Gemma 3 12B IT12BQ8_013.4 GBQ6_K23details
Llama Guard 4 12B12BQ6_K11.5 GBQ4_K_M29details
KORMo Web 11B11BQ8_013.0 GBQ5_K_M24details
Llama 3.2 11B Vision Instruct11BQ8_011.6 GBQ5_K_M28details
Llama 3.2 11B Vision11BQ8_012.4 GBQ5_K_M25details
GLM 4.6V Flash10BQ8_010.3 GBQ8_029details
GLM 4.1V 9B Thinking10BQ8_010.3 GBQ8_029details
Step3 VL 10B10BQ8_09.9 GBQ8_033details
SenseNova Loop 9B Preview9.7BQ8_010.0 GBQ8_030details
Qwen3.5 9B9.7BQ8_010.3 GBQ8_029details
Ornith 1.5 9B9.7BQ8_010.0 GBQ8_030details
SenseNova Loop 9B9.4BQ8_010.0 GBQ8_030details
MiMo V2.6 Distill Qwen 9B9.4BQ8_010.0 GBQ8_030details
GLM 4 9B9.4BQ8_010.3 GBQ8_029details
Apertus v1.5 8B8.9BQ8_010.5 GBQ8_030details
InternLM3 8B Instruct8.8BQ8_09.7 GBQ8_031details
Granite 4.2 8B8.8BQ8_010.5 GBQ6_K31details
Granite 4.1 8B8.8BQ8_010.5 GBQ6_K31details
Granite 4.1 8B Llmon Preview8.8BQ8_010.5 GBQ6_K31details
Qwen3 VL 8B Instruct8.8BQ8_09.9 GBQ8_033details
Molmo2 8B8.7BQ8_09.9 GBQ8_033details
LFM2.5 8B A1B8.5B (1.0B act.)Q8_09.1 GBQ8_0271details
Qwen2.5 VL 7B Instruct8.3BQ8_08.7 GBQ8_036details
UI TARS 1.5 7B8.3BQ8_08.7 GBQ8_036details
olmOCR 2 7B 10258.3BQ8_08.7 GBQ8_036details
olmOCR 7B 0225 Preview8.3BQ8_08.7 GBQ8_036details
UI TARS 7B DPO8.3BQ8_08.7 GBQ8_036details
Qwen3 8B8.2BQ8_09.9 GBQ8_033details
DeepSeek R1 0528 Qwen3 8B8.2BQ8_09.9 GBQ8_033details
Granite 3.3 8B Instruct8.2BQ8_09.8 GBQ8_033details
Granite 3.1 8B Instruct8.2BQ8_09.8 GBQ8_033details
Granite 3.2 8B Instruct8.2BQ8_09.8 GBQ8_033details
Apertus 8B Instruct 25098.1BQ8_09.6 GBQ8_034details
Llama 3.1 8B Instruct8.0BQ8_09.6 GBQ8_034details
Llama 3.1 8B8.0BQ8_09.6 GBQ8_034details
Hermes 3 Llama 3.1 8B8.0BQ8_09.6 GBQ8_034details
Llama Guard 3 8B8.0BQ8_09.6 GBQ8_034details
DeepSeek R1 Distill Llama 8B8.0BQ8_09.6 GBQ8_034details
Meta Llama 3 8B Instruct8.0BQ8_09.6 GBQ8_034details
Meta Llama Guard 2 8B8.0BQ8_09.6 GBQ8_034details
Meta Llama 3.1 8B Instruct8.0BQ8_09.6 GBQ8_034details
Aya Expanse 8B8.0BQ8_09.8 GBQ8_034details
Ling 3.0 Tiny7.9B (1.4B act.)Q8_08.7 GBQ8_0196details
MiMo 7B RL7.8BQ8_09.4 GBQ8_036details
InternLM2.5 7B Chat7.7BQ8_09.2 GBQ8_035details
InternLM2 Chat 7B7.7BQ8_09.2 GBQ8_035details
Qwen2.5 7B Instruct7.6BQ8_08.7 GBQ8_036details
DeepSeek R1 Distill Qwen 7B7.6BQ8_08.7 GBQ8_036details
Qwen2.5 Coder 7B Instruct7.6BQ8_08.7 GBQ8_036details
Olmo 3 7B Instruct7.3BQ8_010.3 GBQ8_037details
Olmo 3 1025 7B7.3BQ8_010.3 GBQ8_037details
Olmo 3 7B Think7.3BQ8_010.3 GBQ8_037details
Mistral 7B v0.17.2BQ8_08.6 GBQ8_037details
Mistral 7B Instruct v0.27.2BQ8_08.6 GBQ8_037details
Mistral 7B Instruct v0.17.2BQ8_08.6 GBQ8_037details
Falcon 7B7.2BQ8_07.7 GBQ8_038details
Falcon 7B Instruct7.2BQ8_07.4 GBQ8_039details
Granite 4.0 H Tiny6.9B (1.5B act.)F1613.6 GBQ8_098details
OLMoE 1B 7B 09246.9B (1.3B act.)Q8_08.4 GBQ8_0211details
DeepSeek Coder 7B Instruct v1.56.9BQ8_011.2 GBno39details
DeepSeek Coder 6.7B Instruct6.7BQ8_011.1 GBno40details
DeepSeek Coder 6.7B Base6.7BQ8_011.1 GBno40details
Llama 2 7B HF6.7BQ8_011.1 GBno40details
Yi 6B 200K6.1BF1612.3 GBQ8_024details
Yi 6B6.1BF1612.3 GBQ8_024details
Molmo2 4B4.9BF1610.0 GBF1633details
WeVisOne VL 4B4.8BF1610.0 GBF1633details
Mage VL Cowork4.7BF1610.0 GBF1633details
Qianfan OCR4.7BF1610.1 GBF1633details
Qwen3.5 4B4.7BF169.2 GBF1633details
SingVL S1 4B4.5BF169.0 GBF1634details
LightOnOCR 3 4B4.5BF169.0 GBF1634details
Intern Decision 4B4.5BF169.0 GBF1634details
Qwen3 VL 4B Instruct4.4BF169.3 GBF1636details
Gemma 3 4B IT4.3BF168.4 GBF1637details
MedGemma 1.5 4B IT4.3BF168.4 GBF1637details
MedGemma 4B IT4.3BF168.4 GBF1637details
TranslateGemma 4B IT4.3BF169.2 GBF1633details
Gemma 3 4B PT4.3BF168.4 GBF1637details
Phi 3.5 Vision Instruct4.1BF1611.2 GBQ4_K_M35details
Qwen3 4B4.0BF169.3 GBF1636details
Qwen3 4B Instruct 25074.0BF169.3 GBF1636details
Granite Vision 4.1 4B4.0BF167.6 GBF1642details
Phi 4 Mini Instruct3.8BF168.9 GBF1638details
Phi 3 Mini 4K Instruct3.8BF1610.6 GBQ5_K_M38details
Phi 3.5 Mini Instruct3.8BF1610.6 GBQ5_K_M38details
Phi 3 Mini 128K Instruct3.8BF1610.6 GBQ5_K_M38details
Qwen2.5 VL 3B Instruct3.8BF168.0 GBF1638details
Granite 4.2 3B3.7BF168.0 GBF1639details
Granite 4.1 3B Llmon Preview3.7BF168.0 GBF1639details
Raon 1 Nano3.5BF167.0 GBF1646details
Granite 4.1 3B3.4BF167.6 GBF1642details
Granite 4.0 Micro3.4BF167.6 GBF1642details
DeepSeek OCR 23.4B (1.2B act.)F166.6 GBF16139details
DeepSeek VL2 Tiny3.4B (1.2B act.)F167.4 GBF16123details
Tiny Aya Global3.3BF167.5 GBF1643details
Unlimited OCR 23.3B (1.1B act.)F166.6 GBF16142details
Unlimited OCR3.3B (1.1B act.)F166.6 GBF16143details
DeepSeek OCR3.3B (1.1B act.)F166.6 GBF16143details
Granite 3.1 3B A800M Instruct3.3B (883M act.)F167.1 GBF16163details
Llama 3.2 3B Instruct3.2BF167.5 GBF1645details
Llama 3.2 3B3.2BF167.5 GBF1645details
Hermes 3 Llama 3.2 3B3.2BF167.5 GBF1645details
d1 3B3.1BF165.8 GBF1653details
LFM2.5 VL 3B3.1BF165.8 GBF1653details
Qwen2.5 3B Instruct3.1BF166.7 GBF1647details
SmolLM3 3B3.1BF166.9 GBF1647details
SmolLM3 3B Base3.1BF166.9 GBF1647details
Phi 22.8BF168.2 GBQ8_052details
LFM2.5 2.6B2.7BF1610.8 GBF1627details
Granite 3.3 2B Instruct2.5BF165.8 GBF1657details
MiniCPM5 2B2.5BF165.7 GBF1657details
WeVisOne VL 2B2.4BF165.4 GBF1671details
Cosmos Reason2 2B2.4BF165.4 GBF1671details
Qwen3.5 2B2.3BF164.6 GBF1674details
SmolVLM Instruct2.2BF165.4 GBF1679details
Qwen3 1.7B2.0BF165.4 GBF1671details
d1 2B1.9BF164.0 GBF1683details
LFM3 VL 2B1.9BF164.0 GBF1683details
DeepSeek R1 Distill Qwen 1.5B1.8BF164.2 GBF1681details
LFM3 1.7B1.7BF167.2 GBF1642details
LFM3 1.7B Base1.7BF164.0 GBF1683details
SmolLM2 1.7B Instruct1.7BF165.2 GBF1684details
SmolLM2 1.7B1.7BF165.2 GBF1684details
SmolLM 1.7B1.7BF165.2 GBF1684details
Granite 4.0 1B1.6BF164.3 GBF1688details
Qwen2.5 1.5B Instruct1.5BF163.8 GBF1693details
OLMo 2 0425 1B1.5BF164.4 GBF1697details
DeepSeek Coder 1.3B Instruct1.3BF164.5 GBF16107details
Llama 3.2 1B1.2BF163.2 GBF16116details
Llama 3.2 1B Instruct1.2BF163.2 GBF16116details
LFM2.5 1.2B Instruct1.2BF165.0 GBF1661details
LFM2 1.2B1.2BF163.0 GBF16123details
HunyuanOCR1.1BF163.5 GBF16129details
PaddleOCR VL 1.6959MF162.5 GBF16150details
Qwen3.5 0.8B873MF162.4 GBF16185details
LightOnOCR 3 0.8B853MF162.4 GBF16190details
FastVLM 0.5B759MF162.2 GBF16190details
Qwen3 0.6B752MF163.0 GBF16191details
GOT OCR2.0716MF162.8 GBF16201details
Nemotron 3.5 Parse631MF161.9 GBF16220details
GOT OCR 2.0 HF561MF162.5 GBF16257details
SmolVLM2 500M Video Instruct507MF161.6 GBF16351details
SmolVLM 500M Instruct507MF161.6 GBF16351details
Qwen2.5 0.5B Instruct494MF161.7 GBF16291details
LFM2.5 VL 450M449MF161.3 GBF16405details
SmolLM2 360M Instruct362MF161.5 GBF16397details
SmolLM2 360M362MF161.5 GBF16397details
LFM2.5 350M354MF161.9 GBF16202details

Bigger models with CPU offload

These popular models don't fit in the RTX 4060 Ti 16GB's VRAM at 4-bit, but run with llama.cpp offload, keeping part of the model in system RAM. Speed range is DDR4-3200 to DDR5-6000 (estimate). Method.

ModelParamsOffload plan
GLM 5.3 Flash321B MoEOffload Q2_K, 128 GB RAM · ~6.8–12 tok/s
gpt-oss 120B117B MoEOffload Q4_K_M, 64 GB RAM · ~13–25 tok/s
DeepSeek V4 Flash 0731304B MoEOffload Q2_K, 128 GB RAM · ~6.8–11 tok/s
DeepSeek V4 Flash291B MoEOffload Q3_K_M, 128 GB RAM · ~6.6–12 tok/s
GLM 4.5 Air110B MoEOffload Q4_K_M, 64 GB RAM · ~5.1–9.1 tok/s
MiniMax M2.5229B MoEOffload Q4_K_M, 128 GB RAM · ~5.3–10 tok/s
Hy3299B MoEOffload Q3_K_M, 128 GB RAM · ~4.0–7.6 tok/s

Comparing cards? The RTX 4060 Ti 16GB (launch MSRP $499) is in best GPUs for local LLMs under $500. See also Mac vs PC.

Just too big at 4-bit

These need up to 35% more memory than the RTX 4060 Ti 16GB has at Q4_K_M and 8K context. A smaller quant or shorter context may work.

ModelQ4_K_M needsShort by
Trinity Mini15.8 GB0.6 GB
GLiDE 27B16.2 GB1.0 GB
MiniMax H3 Image PE16.8 GB1.6 GB
DiffusionGemma 26B A4B IT17.1 GB1.9 GB
Gemma 4 26B A4B IT17.3 GB2.1 GB
Gemma 3 27B IT17.3 GB2.1 GB
Arete 27B17.4 GB2.2 GB
Qwen3.8 27B17.6 GB2.4 GB
Muse Glimmer 30B17.9 GB2.7 GB
Qwen3.5 27B18.1 GB2.9 GB
Qwen3.6 27B18.1 GB2.9 GB
GLM 4.7 Flash18.3 GB3.1 GB
Hy Memory 30B A3B18.5 GB3.3 GB
Qwen3 Coder 30B A3B Instruct18.8 GB3.6 GB
North Mini Code 1.018.8 GB3.6 GB

Questions

What is the biggest model that runs on the RTX 4060 Ti 16GB?

Of the models we track, the largest that fits at 8K context is Maxwell V1 Flash (37B) at Q2_K, needing about 14.0 GB.

How much of the RTX 4060 Ti 16GB's memory can a model use?

About 15.2 GB: we assume 95% of the 16 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.