What can Intel Arc A770 16GB run for vision?

Build: Intel Arc A770 16GB + — + 32 GB RAM (windows)

Memory: 16 GB VRAM + 32 GB system RAM
Runner: llama.cpp (Vulkan)
AnyChatCodingAgentsReasoningVisionLong contextCreative

Runs comfortably
16 models

Ranked by fit for vision use case + predicted speed. Click a row for VRAM breakdown.

#1Moondream 2
1.9B
other
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 2.5 GBHeadroom: 13.5 GB
157
tok/s
Estimated
Weights
1.20 GB
KV cache
0.24 GB
Activations
0.06 GB
Runtime
1.00 GB
Model details →
#2Qwen 2.5-VL 3B
3B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 4.6 GBHeadroom: 11.4 GB
99
tok/s
Estimated
Weights
2.00 GB
KV cache
1.50 GB
Activations
0.11 GB
Runtime
1.00 GB
Model details →
#3Gemma 4 E2B (Effective 2B)
2B
gemma
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 4.3 GBHeadroom: 11.7 GB
ollama run gemma4:e2b
85
tok/s
Estimated
Weights
2.20 GB
KV cache
1.00 GB
Activations
0.12 GB
Runtime
1.00 GB
Model details →
#4Phi-3.5 Vision
4.2B
phi
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 5.7 GBHeadroom: 10.3 GB
71
tok/s
Estimated
Weights
2.50 GB
KV cache
2.10 GB
Activations
0.13 GB
Runtime
1.00 GB
Model details →
#5Qwen 2.5-VL 7B
7B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.1 GBHeadroom: 6.9 GB
43
tok/s
Estimated
Weights
4.40 GB
KV cache
3.50 GB
Activations
0.23 GB
Runtime
1.00 GB
Model details →
#6Qwen 2-VL 7B
7B
qwen
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.3 GBHeadroom: 6.7 GB
43
tok/s
Estimated
Weights
4.60 GB
KV cache
3.50 GB
Activations
0.24 GB
Runtime
1.00 GB
Model details →
#7LLaVA 1.6 Mistral 7B
7B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.2 GBHeadroom: 6.8 GB
43
tok/s
Estimated
Weights
4.50 GB
KV cache
3.50 GB
Activations
0.23 GB
Runtime
1.00 GB
Model details →
#8LLaVA-OneVision 7B
7B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 9.2 GBHeadroom: 6.8 GB
43
tok/s
Estimated
Weights
4.50 GB
KV cache
3.50 GB
Activations
0.23 GB
Runtime
1.00 GB
Model details →
#9Gemma 4 E4B (Effective 4B)
4B
gemma
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 7.6 GBHeadroom: 8.4 GB
ollama run gemma4:e4b
42
tok/s
Estimated
Weights
4.40 GB
KV cache
2.00 GB
Activations
0.23 GB
Runtime
1.00 GB
Model details →
#10Gemma 3 4B
4B
gemma
Commercial OK
Quant: Q8_0Context: 8,192VRAM: 7.6 GBHeadroom: 8.4 GB
ollama run gemma3:4b
42
tok/s
Estimated
Weights
4.40 GB
KV cache
2.00 GB
Activations
0.23 GB
Runtime
1.00 GB
Model details →
#11MiniCPM-V 2.6 8B
8B
minicpm
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 10.3 GBHeadroom: 5.7 GB
37
tok/s
Estimated
Weights
5.00 GB
KV cache
4.00 GB
Activations
0.26 GB
Runtime
1.00 GB
Model details →
#12MiniCPM-V 3 8B
8B
minicpm
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 10.3 GBHeadroom: 5.7 GB
37
tok/s
Estimated
Weights
5.00 GB
KV cache
4.00 GB
Activations
0.26 GB
Runtime
1.00 GB
Model details →

Runs with tradeoffs
12 models

Tight VRAM, partial CPU offload, or context-limited.

Llama 3.2 11B Vision Instruct
11B
llama
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 14.8 GBHeadroom: 1.2 GB
  • • Tight VRAM fit — only 1.2 GB headroom left for context growth
ollama run llama3.2-vision:11b
27
tok/s
Estimated
Weights
7.90 GB
KV cache
5.50 GB
Activations
0.40 GB
Runtime
1.00 GB
Model details →
Llama 3.2 11B Vision
11B
llama
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 13.3 GBHeadroom: 2.7 GB
  • • Tight VRAM fit — only 2.7 GB headroom left for context growth
27
tok/s
Estimated
Weights
6.50 GB
KV cache
5.50 GB
Activations
0.33 GB
Runtime
1.00 GB
Model details →
Gemma 3 12B
12B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 14.7 GBHeadroom: 1.3 GB
  • • Tight VRAM fit — only 1.3 GB headroom left for context growth
ollama run gemma3:12b
25
tok/s
Estimated
Weights
7.30 GB
KV cache
6.00 GB
Activations
0.37 GB
Runtime
1.00 GB
Model details →
Pixtral 12B
12B
mistral
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 14.4 GBHeadroom: 1.6 GB
  • • Tight VRAM fit — only 1.6 GB headroom left for context growth
ollama run pixtral:12b
25
tok/s
Estimated
Weights
7.00 GB
KV cache
6.00 GB
Activations
0.36 GB
Runtime
1.00 GB
Model details →
Phi-4 Multimodal
14B
phi
Commercial OK
Quant: Q4_K_MContext: 2,048VRAM: 12.2 GBHeadroom: 3.8 GB
  • • Tight VRAM fit — only 3.8 GB headroom left for context growth
21
tok/s
Estimated
Weights
9.00 GB
KV cache
1.75 GB
Activations
0.45 GB
Runtime
1.00 GB
Model details →
Gemma 4 26B MoE
26B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 30.8 GBHeadroom: 4.4 GB
  • • Partial CPU offload: ~48% of layers run on CPU
ollama run gemma4:26b-moe
11
tok/s
Estimated
Weights
16.00 GB
KV cache
13.00 GB
Activations
0.81 GB
Runtime
1.00 GB
Model details →
InternVL 2.5 26B
26B
other
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 30.8 GBHeadroom: 4.4 GB
  • • Partial CPU offload: ~48% of layers run on CPU
11
tok/s
Estimated
Weights
16.00 GB
KV cache
13.00 GB
Activations
0.81 GB
Runtime
1.00 GB
Model details →
Gemma 3 27B
27B
gemma
Commercial OK
Quant: Q4_K_MContext: 8,192VRAM: 31.3 GBHeadroom: 3.9 GB
  • • Partial CPU offload: ~49% of layers run on CPU
ollama run gemma3:27b
11
tok/s
Estimated
Weights
16.00 GB
KV cache
13.50 GB
Activations
0.81 GB
Runtime
1.00 GB
Model details →

What if you upgraded?

Hypothetical scenarios. We re-ran the compatibility engine for each.

+32 GB system RAM

Check the current price

Adds 32 GB to your CPU-offload working set. Helps when models don't quite fit in VRAM.

Unlocks: 92 new comfortable, 69 new tradeoff

  • • Qwen 3 0.6B
  • • Llama 3.1 8B Instruct
  • • Qwen 3 8B
  • • Qwen 3 1.7B
Shop this upgrade↗

Add a second Intel Arc A770 16GB

Launch MSRP $349 (2022). Check the current price.

Tensor parallelism splits the model across both cards, effectively doubling VRAM. Bandwidth doesn't double — runs ~1.5× the single-card speed in practice.

Unlocks: 137 new comfortable

  • • Qwen 3 0.6B
  • • Llama 3.1 8B Instruct
  • • Qwen 3 30B-A3B
  • • Qwen 2.5 Coder 32B Instruct
Shop this upgrade↗

Some links above are affiliate links. We may earn a commission at no extra cost to you. How we make money.

Won't run
top 5 popular models

Need more memory than you have. Shown for orientation.

DeepSeek V4 Pro (1.6T MoE)
1600B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

—
Qwen 3.5 235B-A17B (MoE)
397B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

—
Qwen 3 235B-A22B
235B
qwen
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

—
DeepSeek R1 (671B reasoning)
671B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

—
DeepSeek V4 Flash (284B MoE)
284B
deepseek
Commercial OK

Even with CPU offload, needs more memory than your VRAM (16 GB) + 60% of system RAM (19 GB) combined.

—

How to read these numbers

Measured here
Measured here - RunLocalAI ran this exact combo on owner hardware with public evidence.

Source-backed
Source-backed / community - a reproduced public source supports the speed, but it is not labeled as owner-measured.

Extrapolated
Extrapolated - predicted from a measured benchmark on similar-bandwidth hardware.

Estimated
Estimated - formula based on VRAM bandwidth and model architecture; not a benchmark row.

RunLocalAI Will-It-Run Framework →

Want a specific benchmark we don't have? Email Contact support and we'll prioritize it.