What LLMs Can NVIDIA A100 (80GB) Run?

Hardware capability guide for NVIDIA A100 (80GB): 80GB memory with 1940 GB/s bandwidth across single-card and multi-GPU deployments.

NVIDIA A100 (80GB) Hardware Specifications

VRAM / Card
80 GB
Mem Bandwidth
1940 GB/s
FP16 Compute
77.97 TFLOPS
Architecture
Ampere
Release Year
2021
Available VRAM
80 GB
Total Bandwidth
1,940 GB/s
Fits FP16 / FP8
43
Fits w/ INT4
61 / 85
DeepSeek-R1-Distill-Qwen-32B
32B • DeepSeek
Fits FP16+
FP16
68.27 GB
FP8
36 GB
INT4
19.86 GB
~22 tok/s (FP16)Open VRAM Calculator
DeepSeek-R1-Distill-Qwen-14B
14B • DeepSeek
Fits FP16+
FP16
32.27 GB
FP8
18 GB
INT4
10.86 GB
~22 tok/s (FP16)Open VRAM Calculator
DeepSeek-R1-Distill-Qwen-7B
7B • DeepSeek
Fits FP16+
FP16
18.27 GB
FP8
11 GB
INT4
7.36 GB
~22 tok/s (FP16)Open VRAM Calculator
DeepSeek-R1-Distill-Qwen-1.5B
1.5B • DeepSeek
Fits FP16+
FP16
7.27 GB
FP8
5.5 GB
INT4
4.61 GB
~22 tok/s (FP16)Open VRAM Calculator
DeepSeek-R1-Distill-Llama-70B
70B • DeepSeek
Fits FP8+
FP16
144.27 GB
FP8
74 GB
INT4
38.86 GB
~45 tok/s (FP8)Open VRAM Calculator
DeepSeek-R1-Distill-Llama-8B
8B • DeepSeek
Fits FP16+
FP16
20.27 GB
FP8
12 GB
INT4
7.86 GB
~22 tok/s (FP16)Open VRAM Calculator
Llama 4 ScoutMoE 17B
109B • Llama 4
Fits INT4
FP16
220.71 GB
FP8
111.21 GB
INT4
56.46 GB
~194 tok/s (INT4)Open VRAM Calculator
Llama 3 8B
8B • Llama 3
Fits FP16+
FP16
17.81 GB
FP8
9.31 GB
INT4
5.06 GB
~103 tok/s (FP16)Open VRAM Calculator
Llama 3 70B
70B • Llama 3
Fits FP8+
FP16
149.56 GB
FP8
78.31 GB
INT4
42.68 GB
~24 tok/s (FP8)Open VRAM Calculator
Llama 3.1 8B
8B • Llama 3
Fits FP16+
FP16
17.81 GB
FP8
9.31 GB
INT4
5.06 GB
~103 tok/s (FP16)Open VRAM Calculator
Llama 3.1 70B
70B • Llama 3
Fits FP8+
FP16
149.56 GB
FP8
78.31 GB
INT4
42.68 GB
~24 tok/s (FP8)Open VRAM Calculator
Llama 3.2 1B
1B • Llama 3
Fits FP16+
FP16
2.55 GB
FP8
1.43 GB
INT4
0.86 GB
~825 tok/s (FP16)Open VRAM Calculator
Llama 3.2 3B
3B • Llama 3
Fits FP16+
FP16
7.23 GB
FP8
3.79 GB
INT4
2.07 GB
~275 tok/s (FP16)Open VRAM Calculator
Llama 3.2 11B
11B • Llama 3
Fits FP16+
FP16
24.11 GB
FP8
12.61 GB
INT4
6.86 GB
~75 tok/s (FP16)Open VRAM Calculator
Llama 3.2 90B
90B • Llama 3
Fits INT4
FP16
191.57 GB
FP8
100.32 GB
INT4
54.7 GB
~37 tok/s (INT4)Open VRAM Calculator
Qwen3.5-122B-A10BMoE 10B
122B • Qwen 3.5
Fits INT4
FP16
246.01 GB
FP8
123.51 GB
INT4
62.26 GB
~330 tok/s (INT4)Open VRAM Calculator
Qwen3.5-35B-A3BMoE 3B
35B • Qwen 3.5
Fits FP16+
FP16
70.98 GB
FP8
35.66 GB
INT4
18.01 GB
~275 tok/s (FP16)Open VRAM Calculator
Qwen3.5-27B
27B • Qwen 3.5
Fits FP16+
FP16
59.22 GB
FP8
30.97 GB
INT4
16.85 GB
~31 tok/s (FP16)Open VRAM Calculator
Qwen3-Next-80B-A3B-InstructMoE 3B
81B • Qwen 3-Next
Fits INT4
FP16
163.48 GB
FP8
81.91 GB
INT4
41.13 GB
~1099 tok/s (INT4)Open VRAM Calculator
Qwen3-Next-80B-A3B-ThinkingMoE 3B
81B • Qwen 3-Next
Fits INT4
FP16
163.48 GB
FP8
81.91 GB
INT4
41.13 GB
~1099 tok/s (INT4)Open VRAM Calculator
Qwen2.5-Math-72B
72B • Qwen 2
Fits INT4
FP16
153.76 GB
FP8
80.51 GB
INT4
43.88 GB
~46 tok/s (INT4)Open VRAM Calculator
Qwen2.5-72B
72B • Qwen 2
Fits INT4
FP16
153.76 GB
FP8
80.51 GB
INT4
43.88 GB
~46 tok/s (INT4)Open VRAM Calculator
Qwen2-72B
72B • Qwen 2
Fits INT4
FP16
153.76 GB
FP8
80.51 GB
INT4
43.88 GB
~46 tok/s (INT4)Open VRAM Calculator
Qwen QwQ-32B
32B • Qwen QwQ
Fits FP16+
FP16
69.1 GB
FP8
36.16 GB
INT4
19.69 GB
~26 tok/s (FP16)Open VRAM Calculator
Qwen2.5-Coder-32B
32B • Qwen 2
Fits FP16+
FP16
69.1 GB
FP8
36.16 GB
INT4
19.69 GB
~26 tok/s (FP16)Open VRAM Calculator
Qwen3-32B
32B • Qwen 3
Fits FP16+
FP16
69.23 GB
FP8
36.23 GB
INT4
19.73 GB
~26 tok/s (FP16)Open VRAM Calculator
Qwen2.5-32B
32B • Qwen 2
Fits FP16+
FP16
69.1 GB
FP8
36.16 GB
INT4
19.69 GB
~26 tok/s (FP16)Open VRAM Calculator
Qwen2-32B
32B • Qwen 2
Fits FP16+
FP16
69.1 GB
FP8
36.16 GB
INT4
19.69 GB
~26 tok/s (FP16)Open VRAM Calculator
Qwen3-30B-A3BMoE 3B
30B • Qwen 3
Fits FP16+
FP16
61.1 GB
FP8
30.73 GB
INT4
15.54 GB
~275 tok/s (FP16)Open VRAM Calculator
Qwen2.5-14B
14B • Qwen 2
Fits FP16+
FP16
30.66 GB
FP8
16.04 GB
INT4
8.72 GB
~59 tok/s (FP16)Open VRAM Calculator
Qwen3-14B
14B • Qwen 3
Fits FP16+
FP16
30.66 GB
FP8
16.04 GB
INT4
8.72 GB
~59 tok/s (FP16)Open VRAM Calculator
Qwen2-14B
14B • Qwen 2
Fits FP16+
FP16
30.66 GB
FP8
16.04 GB
INT4
8.72 GB
~59 tok/s (FP16)Open VRAM Calculator
Qwen3-8B
8B • Qwen 3
Fits FP16+
FP16
17.93 GB
FP8
9.37 GB
INT4
5.09 GB
~103 tok/s (FP16)Open VRAM Calculator
Qwen2.5-7B
7B • Qwen 2
Fits FP16+
FP16
15.21 GB
FP8
7.96 GB
INT4
4.33 GB
~118 tok/s (FP16)Open VRAM Calculator
Qwen2-7B
7B • Qwen 2
Fits FP16+
FP16
15.21 GB
FP8
7.96 GB
INT4
4.33 GB
~118 tok/s (FP16)Open VRAM Calculator
Qwen3-4B
4B • Qwen 3
Fits FP16+
FP16
9 GB
FP8
4.72 GB
INT4
2.58 GB
~206 tok/s (FP16)Open VRAM Calculator
Qwen2.5-3B
3B • Qwen 2
Fits FP16+
FP16
6.57 GB
FP8
3.46 GB
INT4
1.91 GB
~275 tok/s (FP16)Open VRAM Calculator
Qwen2.5-1.5B
1.5B • Qwen 2
Fits FP16+
FP16
3.49 GB
FP8
1.89 GB
INT4
1.1 GB
~550 tok/s (FP16)Open VRAM Calculator
Qwen2.5-0.5B
0.5B • Qwen 2
Fits FP16+
FP16
1.36 GB
FP8
0.83 GB
INT4
0.57 GB
~1649 tok/s (FP16)Open VRAM Calculator
Qwen3-0.6B
0.6B • Qwen 3
Fits FP16+
FP16
1.94 GB
FP8
1.12 GB
INT4
0.71 GB
~1374 tok/s (FP16)Open VRAM Calculator
Qwen3-1.7B
1.7B • Qwen 3
Fits FP16+
FP16
4.14 GB
FP8
2.22 GB
INT4
1.26 GB
~485 tok/s (FP16)Open VRAM Calculator
Mistral-Large-Instruct-2407
123B • Mistral
Fits INT4
FP16
261.15 GB
FP8
136.77 GB
INT4
74.59 GB
~27 tok/s (INT4)Open VRAM Calculator
Mistral-Nemo-12B
12B • Mistral
Fits FP16+
FP16
26.46 GB
FP8
13.83 GB
INT4
7.52 GB
~69 tok/s (FP16)Open VRAM Calculator
Mistral-7B-v0.3
7B • Mistral
Fits FP16+
FP16
15.71 GB
FP8
8.21 GB
INT4
4.46 GB
~118 tok/s (FP16)Open VRAM Calculator
Mixtral-8x7BMoE 12.9B
46.7B • Mixtral
Fits FP8+
FP16
95.7 GB
FP8
48.5 GB
INT4
24.9 GB
~128 tok/s (FP8)Open VRAM Calculator
Mixtral-8x22BMoE 39B
141B • Mixtral
Fits INT4
FP16
287.68 GB
FP8
145.81 GB
INT4
74.87 GB
~85 tok/s (INT4)Open VRAM Calculator
Gemma-2-27B
27B • Gemma
Fits FP16+
FP16
59.6 GB
FP8
31.16 GB
INT4
16.94 GB
~31 tok/s (FP16)Open VRAM Calculator
Gemma-2-9B
9B • Gemma
Fits FP16+
FP16
21.53 GB
FP8
11.22 GB
INT4
6.06 GB
~92 tok/s (FP16)Open VRAM Calculator
Gemma-2-2B
2B • Gemma
Fits FP16+
FP16
5.11 GB
FP8
2.71 GB
INT4
1.5 GB
~412 tok/s (FP16)Open VRAM Calculator
Phi-3.5-MoEMoE 6.6B
42B • Phi
Fits FP8+
FP16
85.67 GB
FP8
43.17 GB
INT4
21.92 GB
~250 tok/s (FP8)Open VRAM Calculator
Phi-3-14B
14B • Phi
Fits FP16+
FP16
30.97 GB
FP8
16.19 GB
INT4
8.8 GB
~59 tok/s (FP16)Open VRAM Calculator
Phi-3-7B
7B • Phi
Fits FP16+
FP16
15.71 GB
FP8
8.21 GB
INT4
4.46 GB
~118 tok/s (FP16)Open VRAM Calculator
Phi-3-3.8B
3.8B • Phi
Fits FP16+
FP16
11.02 GB
FP8
5.72 GB
INT4
3.07 GB
~217 tok/s (FP16)Open VRAM Calculator
CodeLlama-34B
34B • CodeLlama
Fits FP16+
FP16
72.93 GB
FP8
38.18 GB
INT4
20.8 GB
~24 tok/s (FP16)Open VRAM Calculator
CodeLlama-13B
13B • CodeLlama
Fits FP16+
FP16
33.56 GB
FP8
17.44 GB
INT4
9.37 GB
~63 tok/s (FP16)Open VRAM Calculator
CodeLlama-7B
7B • CodeLlama
Fits FP16+
FP16
18.71 GB
FP8
9.71 GB
INT4
5.21 GB
~118 tok/s (FP16)Open VRAM Calculator
GLM-4.5-AirMoE 12B
106B • GLM-4
Fits INT4
FP16
217.73 GB
FP8
110.48 GB
INT4
56.85 GB
~103 tok/s (INT4)Open VRAM Calculator
GLM-4.5-Air-BaseMoE 12B
106B • GLM-4
Fits INT4
FP16
217.73 GB
FP8
110.48 GB
INT4
56.85 GB
~103 tok/s (INT4)Open VRAM Calculator
GLM-4.7-FlashMoE 3B
30B • GLM-4
Fits FP16+
FP16
66.1 GB
FP8
34.66 GB
INT4
18.94 GB
~26 tok/s (FP16)Open VRAM Calculator
GPT-OSS-120BMoE 5.1B
117B • GPT-OSS
Fits INT4
FP16
235.65 GB
FP8
118.09 GB
INT4
59.31 GB
~647 tok/s (INT4)Open VRAM Calculator
GPT-OSS-20BMoE 3.6B
21B • GPT-OSS
Fits FP16+
FP16
43.15 GB
FP8
21.78 GB
INT4
11.09 GB
~229 tok/s (FP16)Open VRAM Calculator

NVIDIA A100 (80GB) 1x / 2x / 4x Multi-GPU LLM Compatibility Matrix

Precomputed at 8K context, batch size 1 across key benchmark models.

LLM ModelFP8 VRAM Req1× (80GB)2× (160GB)4× (320GB)
DeepSeek-R1(671B)675 GBNeeds 5×Needs 5×Needs 5×
DeepSeek-R1-Distill-Qwen-32B(32B)36 GBFP16 / FP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
Llama 4 Scout(109B)111.21 GBINT4FP8 / INT4FP16 / FP8 / INT4
Llama 3.1 8B(8B)9.31 GBFP16 / FP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
Llama 3.1 70B(70B)78.31 GBFP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
Qwen3-235B-A22B(235B)237.95 GBNeeds 2×INT4FP8 / INT4
Qwen3-32B(32B)36.23 GBFP16 / FP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
Qwen3-14B(14B)16.04 GBFP16 / FP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
Mixtral-8x7B(46.7B)48.5 GBFP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4
GPT-OSS-20B(21B)21.78 GBFP16 / FP8 / INT4FP16 / FP8 / INT4FP16 / FP8 / INT4

Dedicated Compatibility Pages for All 22 GPUs

NVIDIA A100 (80GB) Local LLM Deployment FAQ

tools.gpuCompatibility.gpuDetail.faq.q1▾
With 80GB of memory on a single NVIDIA A100 (80GB), 61 benchmark models in our registry fit cleanly in FP16, FP8, or INT4 at 8K context.
What generation speed can I expect from NVIDIA A100 (80GB)?▾
With 1940 GB/s memory bandwidth and 80GB memory, single-stream decode speed scales inversely with active model weight size read per token.
When should I scale to 2x or 4x NVIDIA A100 (80GB)?▾
Multi-GPU tensor parallelism becomes necessary when model weights plus KV cache exceed 80GB or when serving high-concurrency batches.

Explore Related LLM & GPU Tools

Browse full AI toolkit directory →