What are the best models to run on NVIDIA RTX 3080 Ti (12GB)?

Based on model rankings and 12 GB usable VRAM: Top Overall: Phi-4 Reasoning Plus (INT4) ~52 tok/s. Top Coding: K2 Horizon 7B (INT4) ~125 tok/s. Optimal Daily Driver: Qwen3.5-9B (INT4) ~106 tok/s.

How are VRAM, throughput, and latency calculated?

Calculations are determined using apxml.com modeling, combining LLM architecture analysis with GPU hardware physics. VRAM and Context Scaling: Memory requirements are determined from model architecture, including parameter count, layer configuration, and attention structure. The KV cache footprint scales deterministically based on context length, batch size, and cache precision. Throughput (TPS): Token generation is calculated in the memory-bandwidth-bound decode phase, measuring the time required to stream model weights and KV cache relative to hardware memory bandwidth. Latency (TTFT): Time to first token is calculated from the compute operations required during prompt prefill relative to hardware tensor compute capacity.

How accurate are these estimates?

Calculations are calibrated against empirical ground-truth benchmark runs across diverse architectures and hardware. Because the underlying mechanics are benchmark-calibrated, estimates remain accurate across different configurations without needing to run physical tests for every scenario. VRAM Accuracy: Memory estimates are deterministic and closely match actual allocation. Minor differences typically stay within 3% to 5% due to CUDA runtime overhead or memory allocator fragmentation. Speed and Latency Accuracy: Throughput and TTFT typically vary within 5% to 10% depending on the specific inference runtime, kernel implementations, and system power limits.

Best LLMs to Run on NVIDIA RTX 3080 Ti (12GB)

Rank	Model	Parameters	Precision	TPS (Approx.)	TTFT
#102	Phi-4 Reasoning Plus	14B	INT4	52 tok/s	~1.3s
#121	Qwen3.5-9B	9B	INT4 Q8	106 tok/s 65 tok/s	~742 ms ~811 ms
#127	K2 Horizon 7B	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
#130	Gemma 4 12B	11.95B	INT4	61 tok/s	~1.1s
#136	Qwen3.5-4B	4B	INT4 Q8 FP16	173 tok/s 123 tok/s 75 tok/s	~344 ms ~346 ms ~351 ms
#140	GLM-4V	9B	INT4	80 tok/s	~746 ms
#145	NVIDIA Nemotron 3 Nano 30B-A3B	3.5B (30B active)	INT4 Q8 FP16	39 tok/s 23 tok/s 12 tok/s	~2.3s ~2.3s ~2.6s
#152	Granite 4.2 8B	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#157	Gemma 4 E4B	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#164	Phi-4	14B	INT4	76 tok/s	~1.1s
#176	DeepSeek-R1 14B	14B	INT4	78 tok/s	~1.1s
#177	Ministral 3 14B	14B	INT4	52 tok/s	~1.3s
#180	Gemma 3 12B	12B	INT4	86 tok/s	~983 ms
#185	Ministral 3 8B	8B	INT4 Q8	87 tok/s 56 tok/s	~664 ms ~784 ms
#187	OLMo 3 7B Think	7B	INT4 Q8	94 tok/s 66 tok/s	~585 ms ~639 ms
#188	Qwen3.5-2B	2B	INT4 Q8 FP16	231 tok/s 182 tok/s 124 tok/s	~186 ms ~187 ms ~189 ms
#189	MiniCPM5-2B	2B	INT4 Q8 FP16	231 tok/s 182 tok/s 124 tok/s	~186 ms ~187 ms ~189 ms
#190	Qwen2.5-14B	14B	INT4	76 tok/s	~1.1s
#193	Gemma 4 E2B	5.1B	INT4 Q8 FP16	152 tok/s 104 tok/s 56 tok/s	~431 ms ~434 ms ~516 ms
#195	Ministral 3 3B	3B	INT4 Q8 FP16	146 tok/s 118 tok/s 81 tok/s	~266 ms ~268 ms ~272 ms
#196	DeepSeek-R1 8B	8B	INT4 Q8	117 tok/s 76 tok/s	~661 ms ~666 ms
#198	Gemma 2 9B	9B	INT4 Q8	106 tok/s 65 tok/s	~742 ms ~811 ms
#201	Qwen3-4B	4B	INT4 Q8 FP16	173 tok/s 123 tok/s 75 tok/s	~344 ms ~346 ms ~351 ms
#202	Qwen3-14B	14B	INT4	76 tok/s	~1.1s
#203	Gemma 3 4B	4B	INT4 Q8 FP16	173 tok/s 123 tok/s 75 tok/s	~344 ms ~346 ms ~351 ms
#204	DeepSeek-R1 7B	7B	INT4 Q8	128 tok/s 84 tok/s	~582 ms ~586 ms
#206	Qwen3.5-0.8B	0.8B	INT4 Q8 FP16	298 tok/s 262 tok/s 206 tok/s	~80 ms ~80 ms ~81 ms
#207	Llama 3.1 8B	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#207	Qwen3-8B	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#212	Llama 3 8B	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#213	Qwen3-1.7B	1.7B	INT4 Q8 FP16	244 tok/s 197 tok/s 137 tok/s	~160 ms ~161 ms ~163 ms
#214	Gemma 3 270M	0.27B	INT4 Q8 FP16	325 tok/s 310 tok/s 280 tok/s	~31 ms ~31 ms ~32 ms
#216	Qwen3-0.6B	0.6B	INT4 Q8 FP16	314 tok/s 283 tok/s 232 tok/s	~61 ms ~61 ms ~62 ms
#217	Phi-4-Mini	3.8B	INT4 Q8 FP16	177 tok/s 127 tok/s 78 tok/s	~328 ms ~330 ms ~335 ms
#218	Qwen2.5-7B	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
#220	Ministral-8B-2410	8B	INT4 Q8	115 tok/s 71 tok/s	~662 ms ~723 ms
#221	DeepSeek-R1 1.5B	1.5B	INT4 Q8 FP16	254 tok/s 208 tok/s 148 tok/s	~142 ms ~143 ms ~145 ms
#222	Llama 3.2 3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
#223	Phi-3-mini	3.8B	INT4 Q8 FP16	177 tok/s 127 tok/s 78 tok/s	~328 ms ~330 ms ~335 ms
#224	Gemma 3n E2B IT	6B (2B active)	INT4 Q8	145 tok/s 128 tok/s	~200 ms ~201 ms
#225	Gemma 2 2B	2B	INT4 Q8 FP16	231 tok/s 182 tok/s 124 tok/s	~186 ms ~187 ms ~189 ms
#226	Mistral-7B-Instruct-v0.1	7.3B	INT4 Q8	122 tok/s 80 tok/s	~606 ms ~610 ms
#227	Llama 3.2 1B	1B	INT4 Q8 FP16	282 tok/s 243 tok/s 185 tok/s	~99 ms ~100 ms ~101 ms
#228	Phi-3-medium	14B	INT4	76 tok/s	~1.1s
#230	OLMo 3 7B Instruct	7B	INT4 Q8	94 tok/s 66 tok/s	~585 ms ~639 ms
#231	Qwen2-7B	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
#232	Phi-3-small	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
#233	Mistral-7B-Instruct-v0.2	7.3B	INT4 Q8	122 tok/s 80 tok/s	~606 ms ~610 ms
#234	Gemma 3 1B	1B	INT4 Q8 FP16	282 tok/s 243 tok/s 185 tok/s	~99 ms ~100 ms ~101 ms
#236	Gemma 1 7B	7B	INT4 Q8	94 tok/s 66 tok/s	~585 ms ~639 ms
#237	Gemma 1 2B	2B	INT4 Q8 FP16	241 tok/s 188 tok/s 126 tok/s	~186 ms ~187 ms ~189 ms
#238	ChatGLM3-6B	6B	INT4 Q8	103 tok/s 77 tok/s	~505 ms ~509 ms
#239	ChatGLM2-6B	6B	INT4 Q8	103 tok/s 77 tok/s	~505 ms ~509 ms
#240	ChatGLM-6B	6B	INT4 Q8	103 tok/s 77 tok/s	~505 ms ~509 ms
Unranked Models (Sorted by Speed)
-	DINOv3 ViT-L/16	0.3B	INT4 Q8 FP16	340 tok/s 321 tok/s 286 tok/s	~34 ms ~34 ms ~34 ms
-	Qwen2-0.5B	0.5B	INT4 Q8 FP16	322 tok/s 295 tok/s 248 tok/s	~51 ms ~52 ms ~52 ms
-	Qwen2.5-0.5B	0.5B	INT4 Q8 FP16	322 tok/s 295 tok/s 248 tok/s	~51 ms ~52 ms ~52 ms
-	ERNIE-4.5-0.3B	0.3B	INT4 Q8 FP16	319 tok/s 304 tok/s 272 tok/s	~34 ms ~34 ms ~34 ms
-	ERNIE-4.5-0.3B-Base	0.3B	INT4 Q8 FP16	319 tok/s 304 tok/s 272 tok/s	~34 ms ~34 ms ~34 ms
-	Liquid D1-Omni 600M	0.6B	INT4 Q8 FP16	314 tok/s 283 tok/s 232 tok/s	~61 ms ~61 ms ~62 ms
-	Falcon-1B	1B	INT4 Q8 FP16	291 tok/s 249 tok/s 188 tok/s	~99 ms ~100 ms ~101 ms
-	Falcon3-1B	1B	INT4 Q8 FP16	282 tok/s 243 tok/s 185 tok/s	~99 ms ~100 ms ~101 ms
-	MiniCPM-V 4.7 1B	1B	INT4 Q8 FP16	282 tok/s 243 tok/s 185 tok/s	~99 ms ~100 ms ~101 ms
-	Qwen2-1.5B	1.5B	INT4 Q8 FP16	254 tok/s 208 tok/s 148 tok/s	~142 ms ~143 ms ~145 ms
-	Qwen2.5-1.5B	1.5B	INT4 Q8 FP16	254 tok/s 208 tok/s 148 tok/s	~142 ms ~143 ms ~145 ms
-	Spark X2.5 1.7B	1.7B	INT4 Q8 FP16	244 tok/s 197 tok/s 137 tok/s	~160 ms ~161 ms ~163 ms
-	Hy-MT2 1.8B	1.8B	INT4 Q8 FP16	240 tok/s 192 tok/s 133 tok/s	~168 ms ~169 ms ~171 ms
-	Falcon-3B	3B	INT4 Q8 FP16	206 tok/s 151 tok/s 95 tok/s	~264 ms ~266 ms ~270 ms
-	CroissantLLM Base	1.3B	INT4 Q8 FP16	203 tok/s 179 tok/s 138 tok/s	~126 ms ~126 ms ~128 ms
-	Phi-1	1.3B	INT4 Q8 FP16	203 tok/s 179 tok/s 138 tok/s	~126 ms ~126 ms ~128 ms
-	Phi-1.5	1.3B	INT4 Q8 FP16	203 tok/s 179 tok/s 138 tok/s	~126 ms ~126 ms ~128 ms
-	Mellum 2.1 12B A2.5B Thinking	12B (2.5B active)	INT4	200 tok/s	~258 ms
-	DeepSeek-R1 3B	3B	INT4 Q8 FP16	199 tok/s 148 tok/s 94 tok/s	~264 ms ~266 ms ~270 ms
-	Falcon3-3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	LFM2-VL 3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	Liquid D1 3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	Liquid D1-VL 3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	Liquid LFM2-VL 3B D1	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	Qwen2.5-3B	3B	INT4 Q8 FP16	198 tok/s 147 tok/s 94 tok/s	~265 ms ~266 ms ~270 ms
-	FrogNano 4B	4B	INT4 Q8 FP16	173 tok/s 123 tok/s 75 tok/s	~344 ms ~346 ms ~351 ms
-	Spark X2.5 4B	4B	INT4 Q8 FP16	173 tok/s 123 tok/s 75 tok/s	~344 ms ~346 ms ~351 ms
-	EmbeddingGemma 2	2B	INT4 Q8 FP16	167 tok/s 143 tok/s 104 tok/s	~187 ms ~188 ms ~191 ms
-	Phi-2	2.7B	INT4 Q8 FP16	151 tok/s 124 tok/s 87 tok/s	~243 ms ~244 ms ~247 ms
-	MaLLaM-3B	3B	INT4 Q8 FP16	146 tok/s 118 tok/s 81 tok/s	~266 ms ~268 ms ~272 ms
-	SmolLM3 3B	3B	INT4 Q8 FP16	146 tok/s 118 tok/s 81 tok/s	~266 ms ~268 ms ~272 ms
-	Falcon-7B	7B	INT4 Q8	130 tok/s 85 tok/s	~582 ms ~586 ms
-	Falcon3-7B	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
-	Hy-MT2 7B	7B	INT4 Q8	125 tok/s 83 tok/s	~582 ms ~586 ms
-	Mistral-7B-v0.1	7.3B	INT4 Q8	122 tok/s 80 tok/s	~606 ms ~610 ms
-	Youtu Parsing Omni	8B	INT4 Q8	117 tok/s 76 tok/s	~661 ms ~666 ms
-	LensVLM 9B	9B	INT4 Q8	106 tok/s 65 tok/s	~742 ms ~811 ms
-	MiMo V2.6 Distill Qwen 9B	9B	INT4 Q8	106 tok/s 65 tok/s	~742 ms ~811 ms
-	ChatGLM3-6B-32K	6B	INT4 Q8	103 tok/s 77 tok/s	~505 ms ~509 ms
-	Yi-6B	6B	INT4 Q8	103 tok/s 77 tok/s	~505 ms ~509 ms
-	Falcon3-10B	10B	INT4 Q8	98 tok/s 56 tok/s	~822 ms ~971 ms
-	Falcon2-11B	11B	INT4 Q8	95 tok/s 53 tok/s	~901 ms ~1.1s
-	Hunyuan Lite	7B	INT4 Q8	94 tok/s 66 tok/s	~585 ms ~639 ms
-	OLMo 3 7B Base	7B	INT4 Q8	94 tok/s 66 tok/s	~585 ms ~639 ms
-	SEA-LION-7B	7.1B	INT4 Q8	93 tok/s 65 tok/s	~593 ms ~647 ms
-	SEA-LION-7B-Instruct	7.1B	INT4 Q8	93 tok/s 65 tok/s	~593 ms ~647 ms
-	Sahabat-AI-Llama3-8B-Instruct	8B	INT4 Q8	87 tok/s 56 tok/s	~664 ms ~784 ms
-	Typhoon-2-8B	8B	INT4 Q8	87 tok/s 56 tok/s	~664 ms ~784 ms
-	GLM-4-9B	9B	INT4	80 tok/s	~746 ms
-	GLM-4-9B-Chat	9B	INT4	80 tok/s	~746 ms
-	GLM-4-9B-Chat-1M	9B	INT4	80 tok/s	~746 ms
-	Yi-9B	9B	INT4	80 tok/s	~746 ms
-	Sahabat-AI-Gemma2-9B	9.2B	INT4	78 tok/s	~761 ms
-	Sahabat-AI-Gemma2-9B-Instruct	9.2B	INT4	78 tok/s	~761 ms
-	GreenMind-14B-R1	14B	INT4	52 tok/s	~1.3s

NVIDIA RTX 3080 Ti (12GB)

VRAM Scaling by Context Length

Best Models to Run on NVIDIA RTX 3080 Ti (12GB)

Inference Capacity

Fine-Tuning Capacity

Workload Recommendations

Frequently Asked Questions