Desktop GPU · NVIDIA

What AI models can a GeForce RTX 4060 Ti 16GB run?

16 GB covers almost everything people actually use day to day, including the 24B to 32B class that most reviewers consider the practical ceiling for a single consumer card. Bandwidth is the weak point at 288 GB/s. Models fit, then generate slowly, because every token means reading the whole active model out of memory.

The cheapest way to 16 GB on a new card, and the slowest 16 GB card there is. The narrow 128-bit bus holds it to 288 GB/s, so models that fit it run at less than half the speed of wider cards with the same memory. Buy it if the model must fit and speed is secondary.

Memory16 GB

Bandwidth288 GB/s

Usable for a model15.2 GB

Runtime backendCUDA

ArchitectureAda Lovelace

Power165 W

The short answer

Assuming an 8k context window and default settings, these are the models worth downloading first.

Best all-rounder

GPT OSS 20B

MXFP4 · 11.55 GB · about 51.5 tokens/s

Best for code

Qwen2.5 Coder 14B Instruct

Q6_K · 12.63 GB · about 17.4 tokens/s

Largest that still runs well

GLM 4.7 Flash

31.22B parameters · Q3_K_M · about 59.6 tokens/s

Fastest useful answer

LFM2.5 8B A1B

about 74.6 tokens/s · 8.81 GB

See how fast it feels

GeForce RTX 4060 Ti 16GB running GPT OSS 20B at MXFP4

Wait for the first word128 ms

Then writes at51.5 tok/s

Whole answer3.5 s

YouWhy does my model use more memory when the conversation gets longer?

Model

Simulated from our estimate at a 512-token question, not a recording. Assumes nothing else is competing for the GPU.

Where it sits

With 16 GB it belongs to the 16 GB tier. The 16 GB guide explains what that tier runs and where it runs out, and the bandwidth guide explains why 288 GB/s decides how fast it answers.

Our indicative price for this card is $450 new, reviewed 2026-09-01. That is 3.6 GB of memory and 64 GB/s of bandwidth for every $100. Prices move, so treat these as a place to start a search and not a quote.

Devices in a similar price band, fastest memory first:

Device	Price	Memory	Bandwidth	Best all-round pick
GeForce RTX 3080 Ti	$450 used	12 GB	912 GB/s	Gemma 4 12B Instruct, 64.2 t/s
GeForce RTX 3080 12GB	$400 used	12 GB	912 GB/s	Gemma 4 12B Instruct, 64.2 t/s
GeForce RTX 3080 10GB	$350 used	10 GB	760 GB/s	Qwen3.5 9B, 81.1 t/s
GeForce RTX 5070	$560	12 GB	672 GB/s	Gemma 4 12B Instruct, 47.3 t/s

Can an RTX 4060 Ti 16GB run it?

A direct answer for the models people ask about most, each with the settings we would use.

Qwen2.5 7B Instruct Llama 3.1 8B Instruct Qwen3 8B Qwen3.5 9B Gemma 4 12B Instruct Qwen3 14B Phi 4 GPT OSS 20B Qwen3.5 27B Gemma 4 31B Instruct Qwen3 32B Qwen3 30B A3B GPT OSS 120B

Every model, scored on this device

Each row uses the highest-quality quantization that both fits and stays conversational. Speed is a single-stream estimate at 8k context.

Model	Params	Verdict	Download	Memory used	Speed	Max context
GPT OSS 20B GPT-OSS	20.91B (4.18B active)	Runs great	MXFP4	11.55 GB	51.5 t/s	128k
Qwen3.5 27B Qwen	27.78B	Runs well	IQ4_XS	14.2 GB	15.4 t/s	16k
Qwen3.5 9B Qwen	9.65B	Runs well	Q8_0	9.25 GB	24 t/s	128k
Qwen3 14B Qwen	14.77B	Runs well	Q6_K	12.38 GB	17.8 t/s	16k
Gemma 4 26B A4B Instruct Gemma	25.81B (3.74B active)	Runs great	IQ4_XS	13.96 GB	46 t/s	16k
Gemma 4 12B Instruct Gemma	11.96B	Runs well	Q8_0	13.55 GB	16.1 t/s	16k
Qwen2.5 14B Instruct Qwen	14.77B	Runs well	Q6_K	12.63 GB	17.4 t/s	16k
Qwen2.5 Coder 14B Instruct Qwen	14.77B	Runs well	Q6_K	12.63 GB	17.4 t/s	16k
Gemma 4 E4B Instruct Gemma	8B	Runs great	Q8_0	8.42 GB	26.3 t/s	128k
Phi 4 Phi	14.66B	Runs well	Q6_K	12.81 GB	17.1 t/s	16k
Granite 4.1 8B Granite	8.79B	Runs well	Q8_0	10.35 GB	21.3 t/s	32k
Qwen3 8B Qwen	8.19B	Runs well	Q8_0	9.02 GB	24.6 t/s	32k
Gemma 3 27B Instruct Gemma	27.43B	Just fits	IQ4_XS	15.11 GB	14.5 t/s	8k
Gemma 3 12B Instruct Gemma	12.19B	Runs well	Q8_0	13.27 GB	16.5 t/s	32k
Qwen2.5 VL 7B Instruct Qwen	8.29B	Runs great	Q8_0	8.49 GB	26.2 t/s	64k
GLM 4.7 Flash GLM	31.22B (3.66B active)	Just fits	Q3_K_M	14.82 GB	59.6 t/s	8k
Mistral Small 24B Instruct 2501 Mistral	23.57B	Just fits	Q4_K_M	14.57 GB	15 t/s	8k
Qwen2.5 7B Instruct Qwen	7.62B	Runs great	Q8_0	7.83 GB	28.5 t/s	32k
Llama 3.1 8B Instruct Llama	8.03B	Runs great	Q8_0	8.83 GB	25.1 t/s	32k
Gemma 4 31B Instruct Gemma	31.27B	Fits only at 2 to 3 bits	IQ3_XXS !	14.87 GB	14.7 t/s	8k
Mistral Nemo Instruct 2407 Mistral	12.25B	Runs well	Q8_0	13.15 GB	16.7 t/s	16k
Qwen3 30B A3B Qwen	30.53B (3.34B active)	Just fits	Q3_K_M	14.84 GB	54.5 t/s	8k
Qwen3 32B Qwen	32.76B	Fits only at 2 to 3 bits	IQ3_XXS !	13.83 GB	15.8 t/s	8k
DeepSeek Coder V2 Lite Instruct DeepSeek	15.71B (2.74B active)	Runs great	Q6_K	12.41 GB	53.4 t/s	64k
Codestral 22B V0.1 Mistral	22.25B	Just fits	Q4_K_M	14.62 GB	15 t/s	8k
Gemma 2 9B Instruct Gemma	9.24B	Runs well	Q8_0	11.5 GB	19.1 t/s	8k
Qwen2.5 Coder 7B Instruct Qwen	7.62B	Runs great	Q8_0	7.83 GB	28.5 t/s	32k
Qwen3 Coder 30B A3B Instruct Qwen	30.53B (3.34B active)	Just fits	Q3_K_M	14.84 GB	54.5 t/s	8k
OLMo 3 7B Instruct OLMo	7.3B	Runs well	Q8_0	9.72 GB	22.7 t/s	32k
Granite 4.1 30B Granite	28.87B	Fits only at 2 to 3 bits	IQ3_XXS !	12.69 GB	17.2 t/s	16k

Show the other 56 models

Internlm3 8B Instruct	8.8B	Runs well	Q8_0	8.97 GB	24.7 t/s	32k
Mistral 7B Instruct V0.3	7.25B	Runs great	Q8_0	8.45 GB	26.4 t/s	32k
Qwen2.5 32B Instruct	32.76B	Fits only at 2 to 3 bits	IQ3_XXS	13.83 GB	15.8 t/s	8k
Gemma 4 E2B Instruct	5.12B	Runs great	Q8_0	5.44 GB	41.4 t/s	128k
Qwen2.5 Coder 32B Instruct	32.76B	Fits only at 2 to 3 bits	IQ3_XXS	13.83 GB	15.8 t/s	8k
Qwen3.5 4B	4.66B	Runs great	Q8_0	5.27 GB	43.2 t/s	256k
Mistral 7B Instruct V0.2	7.24B	Runs great	Q8_0	8.44 GB	26.4 t/s	32k
DeepSeek Coder 7B Instruct V1.5	6.91B	Runs well	Q8_0	10.58 GB	20.8 t/s	4k
DeepSeek Coder 6.7B Instruct	6.74B	Runs well	Q8_0	10.94 GB	20.1 t/s	16k
LFM2.5 8B A1B	8.47B	Runs great	Q8_0	8.81 GB	74.6 t/s	64k
Qwen3 4B	4.02B	Runs great	Q8_0	5.46 GB	41.5 t/s	32k
Gemma 3 4B Instruct	4.3B	Runs great	Q8_0	4.91 GB	46.6 t/s	128k
Phi 4 Mini Instruct	3.84B	Runs great	Q8_0	5.17 GB	44.1 t/s	64k
Granite 4.1 3B	3.4B	Runs great	Q8_0	4.34 GB	53.1 t/s	128k
Qwen2.5 3B Instruct	3.09B	Runs great	Q8_0	3.68 GB	63.5 t/s	32k
Phi 3 Mini 4k Instruct	3.82B	Runs great	Q8_0	4.8 GB	47.8 t/s	4k
Llama 3.2 3B Instruct	3.21B	Runs great	Q8_0	4.42 GB	52.3 t/s	64k
LFM2.5 2.6B	2.7B	Runs great	Q8_0	3.14 GB	75.8 t/s	128k
Qwen3.5 2B	2.27B	Runs great	Q8_0	2.7 GB	89.9 t/s	256k
Mixtral 8x7B Instruct V0.1	46.7B	Fits only at 2 to 3 bits	IQ2_XXS	12.57 GB	30.1 t/s	16k
Gemma 2 2B Instruct	2.61B	Runs great	Q8_0	3.54 GB	66.4 t/s	8k
Starcoder2 3B	3.03B	Runs great	Q8_0	3.34 GB	71.4 t/s	16k
Qwen3 1.7B	2.03B	Runs great	Q8_0	3.22 GB	73.5 t/s	32k
DeepSeek R1 Distill Qwen 1.5B	1.78B	Runs great	Q8_0	2.07 GB	121.1 t/s	128k
Qwen2.5 1.5B Instruct	1.54B	Runs great	Q8_0	2.06 GB	121.6 t/s	32k
SmolLM2 1.7B	1.71B	Runs great	Q8_0	3.53 GB	66.4 t/s	8k
OLMo 2 0425 1B	1.48B	Runs great	Q8_0	2.6 GB	93.7 t/s	4k
Llama 3.2 1B Instruct	1.24B	Runs great	Q8_0	1.82 GB	143.5 t/s	128k
LFM2.5 1.2B Instruct	1.17B	Runs great	Q8_0	1.46 GB	189.4 t/s	64k
Gemma 3 1B Instruct	1B	Runs great	Q8_0	1.35 GB	204 t/s	32k
Qwen3.5 0.8B	0.87B	Runs great	Q8_0	1.28 GB	217.8 t/s	256k
Qwen3 0.6B	0.75B	Runs great	Q8_0	1.92 GB	131 t/s	32k
Qwen2.5 0.5B Instruct	0.49B	Runs great	Q8_0	0.88 GB	366.2 t/s	32k
LFM2.5 350M	0.35B	Runs great	Q8_0	0.68 GB	567 t/s	64k
Qwen3.5 35B A3B	35.95B	Partial offload	n/a	20.48 GB	59.3 t/s	n/a
Qwen3 Coder Next	79.67B	Partial offload	n/a	45.22 GB	33.2 t/s	n/a
Qwen3 Next 80B A3B Instruct	81.32B	Partial offload	n/a	46.15 GB	32.7 t/s	n/a
GPT OSS 120B	116.83B	Partial offload	n/a	59.4 GB	17.1 t/s	n/a
Phi 3.5 MoE Instruct	41.87B	Partial offload	n/a	24.87 GB	14.6 t/s	n/a
Qwen3.5 122B A10B	125.09B	Partial offload	n/a	70.61 GB	12.4 t/s	n/a
GLM 4.5 Air	110.47B	Partial offload	n/a	63.61 GB	6.8 t/s	n/a
MiniMax M2.5	228.7B	Partial offload	n/a	130.56 GB	6.5 t/s	n/a
MiniMax M2.7	228.69B	Partial offload	n/a	130.55 GB	6.5 t/s	n/a
Qwen3 235B A22B	235.09B	Partial offload	n/a	133.71 GB	3.4 t/s	n/a
MiniMax M3	427.04B	Partial offload	n/a	240.84 GB	3.3 t/s	n/a
GLM 4.5	358.34B	Partial offload	n/a	204.37 GB	2.4 t/s	n/a
Kimi K2.6	1026.88B	Partial offload	n/a	577.78 GB	2.1 t/s	n/a
Kimi K2 Instruct	1026.41B	Partial offload	n/a	577.52 GB	2.1 t/s	n/a
DeepSeek V3.2	685.4B	Partial offload	n/a	385.91 GB	2.1 t/s	n/a
DeepSeek R1	684.53B	Partial offload	n/a	385.42 GB	2.1 t/s	n/a
GLM 5.1	753.86B	Partial offload	n/a	424.51 GB	2 t/s	n/a
GLM 5.2	753.33B	Partial offload	n/a	424.21 GB	1 t/s	n/a
Qwen2.5 VL 72B Instruct	73.41B	Partial offload	n/a	43.62 GB	1 t/s	n/a
Qwen2.5 72B Instruct	72.71B	Partial offload	n/a	43.22 GB	1 t/s	n/a
Llama 3.3 70B Instruct	70.55B	Partial offload	n/a	42.12 GB	1 t/s	n/a
Llama 3 3 Nemotron Super 49B V1	49.87B	Partial offload	n/a	47.99 GB	0.8 t/s	n/a

What it will not run

0 of the 86 architectures we track are out of reach here, even at 2-bit precision. Another 22 run by splitting the model between the card and system RAM, which works but drops generation to single digits for most dense models.

If you want more headroom

The next steps up in memory, in order. More memory changes which models load at all; more bandwidth changes how fast they answer, so the two are worth weighing separately.

Device	Memory	Bandwidth	Models that fit
Radeon RX 7900 XT	20 GB	800 GB/s	64	Check price
GeForce RTX 4090	24 GB	1008 GB/s	69	Check price
GeForce RTX 3090 Ti	24 GB	1008 GB/s	69	Check price
Radeon RX 7900 XTX	24 GB	960 GB/s	69	Check price

Price links go to an Amazon search for the model name and are affiliate links: if you buy through one, we earn a commission at no cost to you. We do not take payment for placement, and the ordering above is by memory capacity alone.