L
localai.computer
ModelsGPUsSystemsAI SetupsBuildsOpenClawMethodology

Resources

  • Methodology
  • Submit Benchmark
  • About

Browse

  • AI Models
  • GPUs
  • PC Builds

Guides

  • OpenClaw Guide
  • How-To Guides

Legal

  • Privacy
  • Terms
  • Contact

© 2025 localai.computer. Hardware recommendations for running AI models locally.

ℹ️We earn from qualifying purchases through affiliate links at no extra cost to you. This supports our free content and research.

  1. Home
  2. GPUs
  3. RX 9070

Quick Answer: RX 9070 offers 16GB VRAM and starts around current market pricing. It delivers approximately 115 tokens/sec on deepseek-ai/DeepSeek-OCR-2. It typically draws 220W under load.

RX 9070

Check availability
By AMDReleased 2025-01MSRP $499.00

This GPU offers reliable throughput for local AI workloads. Pair it with the right model quantization to hit your desired tokens/sec, and monitor prices below to catch the best deal.

Search on AmazonView Benchmarks
Specs snapshot
Key hardware metrics for AI workloads.
VRAM16GB
Cores3,584
TDP220W
ArchitectureRDNA 4
Key Takeaways
  • 16GB VRAM - runs models up to ~40B parameters
  • Entry-level to mid-range performance
  • Efficient (220W) - works with standard PSU configurations
  • Strong price-to-VRAM value

What this means for you

With 16GB VRAM, RX 9070 can run models up to approximately 40B parameters using 4-bit quantization. This handles most popular models including Llama 3 70B, Mistral 7B, and larger.

Who should buy

  • Running 13B-34B parameter models
  • Stable Diffusion XL at high resolution
  • Enthusiast-level local AI experimentation

Looking to upgrade?

Consider RTX 4090 — Double the VRAM for larger models.

Where to Buy

Buy directly on Amazon with fast shipping and reliable customer service.

No purchase links available yet. Try the Amazon search results to find this GPU.
Complete Your Build

Essential accessories to pair with RX 9070

Corsair RM750x (2025) 750W
Minimum 750W recommended for RTX 40 series
$119
Find on Amazon
Corsair Vengeance 32GB DDR5-6000
32GB ideal for AI workloads
$129
Find on Amazon
Noctua NF-A12x25 PWM
Quiet and efficient cooling
$35
Find on Amazon
Thermal Grizzly Kryonaut
Premium thermal paste for optimal cooling
$15
Find on Amazon

Total Bundle Price

All items from Amazon

$797
Individual: $797
Find All on AmazonMore GPUs

💡 Not ready to buy? Try cloud GPUs first

Test RX 9070 performance in the cloud before investing in hardware. Pay by the hour with no commitment.

Vast.aifrom $0.20/hrRunPodfrom $0.30/hrLambda Labsenterprise-grade

AI benchmarks

ModelQuantizationTokens/secVRAM used
deepseek-ai/DeepSeek-OCR-2Q4
115.05 tok/sEstimated

Auto-generated benchmark

2GB
nari-labs/Dia2-2BQ4
114.63 tok/sEstimated

Auto-generated benchmark

2GB
unsloth/Llama-3.2-1B-InstructQ4
113.89 tok/sEstimated

Auto-generated benchmark

1GB
LiquidAI/LFM2-1.2BQ4
112.55 tok/sEstimated

Auto-generated benchmark

1GB
WeiboAI/VibeThinker-1.5BQ4
112.36 tok/sEstimated

Auto-generated benchmark

1GB
ibm-research/PowerMoE-3bQ4
112.33 tok/sEstimated

Auto-generated benchmark

2GB
deepseek-ai/DeepSeek-OCRQ4
111.55 tok/sEstimated

Auto-generated benchmark

2GB
meta-llama/Llama-3.2-3B-InstructQ4
111.43 tok/sEstimated

Auto-generated benchmark

2GB
google/gemma-2bQ4
110.88 tok/sEstimated

Auto-generated benchmark

1GB
facebook/sam3Q4
110.50 tok/sEstimated

Auto-generated benchmark

1GB
Qwen/Qwen2.5-3B-InstructQ4
110.39 tok/sEstimated

Auto-generated benchmark

2GB
ibm-granite/granite-3.3-2b-instructQ4
109.60 tok/sEstimated

Auto-generated benchmark

1GB
nineninesix/kani-tts-2-enQ4
108.96 tok/sEstimated

Auto-generated benchmark

1GB
unsloth/gemma-3-1b-itQ4
108.79 tok/sEstimated

Auto-generated benchmark

1GB
allenai/OLMo-2-0425-1BQ4
108.38 tok/sEstimated

Auto-generated benchmark

1GB
google/gemma-2-2b-itQ4
106.14 tok/sEstimated

Auto-generated benchmark

1GB
TinyLlama/TinyLlama-1.1B-Chat-v1.0Q4
105.12 tok/sEstimated

Auto-generated benchmark

1GB
meta-llama/Llama-3.2-1BQ4
102.76 tok/sEstimated

Auto-generated benchmark

1GB
Qwen/Qwen2.5-3BQ4
102.56 tok/sEstimated

Auto-generated benchmark

2GB
google-bert/bert-base-uncasedQ4
102.04 tok/sEstimated

Auto-generated benchmark

1GB
deepseek-ai/deepseek-coder-1.3b-instructQ4
101.88 tok/sEstimated

Auto-generated benchmark

2GB
meta-llama/Llama-3.2-1B-InstructQ4
101.82 tok/sEstimated

Auto-generated benchmark

1GB
google-t5/t5-3bQ4
101.16 tok/sEstimated

Auto-generated benchmark

2GB
bigcode/starcoder2-3bQ4
100.91 tok/sEstimated

Auto-generated benchmark

2GB
meta-llama/Llama-3.2-3BQ4
100.25 tok/sEstimated

Auto-generated benchmark

2GB
inference-net/Schematron-3BQ4
100.22 tok/sEstimated

Auto-generated benchmark

2GB
google/embeddinggemma-300mQ4
100.15 tok/sEstimated

Auto-generated benchmark

1GB
apple/OpenELM-1_1B-InstructQ4
99.78 tok/sEstimated

Auto-generated benchmark

1GB
google/gemma-3-1b-itQ4
98.40 tok/sEstimated

Auto-generated benchmark

1GB
meta-llama/Llama-Guard-3-1BQ4
97.95 tok/sEstimated

Auto-generated benchmark

1GB
context-labs/meta-llama-Llama-3.2-3B-Instruct-FP16Q4
96.24 tok/sEstimated

Auto-generated benchmark

2GB
tencent/HunyuanOCRQ4
95.78 tok/sEstimated

Auto-generated benchmark

1GB
meta-llama/Llama-2-7b-hfQ4
95.62 tok/sEstimated

Auto-generated benchmark

4GB
Qwen/Qwen2-0.5B-InstructQ4
95.62 tok/sEstimated

Auto-generated benchmark

3GB
microsoft/Phi-4-multimodal-instructQ4
95.54 tok/sEstimated

Auto-generated benchmark

4GB
Qwen/Qwen3-8B-BaseQ4
95.47 tok/sEstimated

Auto-generated benchmark

4GB
Qwen/Qwen3-Embedding-4BQ4
95.44 tok/sEstimated

Auto-generated benchmark

2GB
tencent/HunyuanVideo-1.5Q4
95.42 tok/sEstimated

Auto-generated benchmark

4GB
Qwen/Qwen3-4B-Instruct-2507Q4
95.38 tok/sEstimated

Auto-generated benchmark

2GB
Qwen/Qwen3-ASR-1.7BQ4
95.34 tok/sEstimated

Auto-generated benchmark

2GB
microsoft/Phi-3.5-mini-instructQ4
95.29 tok/sEstimated

Auto-generated benchmark

2GB
ibm-granite/granite-3.3-8b-instructQ4
95.10 tok/sEstimated

Auto-generated benchmark

4GB
GSAI-ML/LLaDA-8B-InstructQ4
94.84 tok/sEstimated

Auto-generated benchmark

4GB
deepseek-ai/DeepSeek-R1-0528Q4
94.80 tok/sEstimated

Auto-generated benchmark

4GB
zai-org/GLM-4.5-AirQ4
94.53 tok/sEstimated

Auto-generated benchmark

4GB
microsoft/DialoGPT-smallQ4
94.52 tok/sEstimated

Auto-generated benchmark

4GB
unsloth/Llama-3.2-3B-InstructQ4
94.51 tok/sEstimated

Auto-generated benchmark

2GB
openai-community/gpt2Q4
94.40 tok/sEstimated

Auto-generated benchmark

4GB
microsoft/DialoGPT-mediumQ4
94.35 tok/sEstimated

Auto-generated benchmark

4GB
liuhaotian/llava-v1.5-7bQ4
94.34 tok/sEstimated

Auto-generated benchmark

4GB
deepseek-ai/DeepSeek-OCR-2
Q4
2GB
115.05 tok/sEstimated
Auto-generated benchmark
nari-labs/Dia2-2B
Q4
2GB
114.63 tok/sEstimated
Auto-generated benchmark
unsloth/Llama-3.2-1B-Instruct
Q4
1GB
113.89 tok/sEstimated
Auto-generated benchmark
LiquidAI/LFM2-1.2B
Q4
1GB
112.55 tok/sEstimated
Auto-generated benchmark
WeiboAI/VibeThinker-1.5B
Q4
1GB
112.36 tok/sEstimated
Auto-generated benchmark
ibm-research/PowerMoE-3b
Q4
2GB
112.33 tok/sEstimated
Auto-generated benchmark
deepseek-ai/DeepSeek-OCR
Q4
2GB
111.55 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-3.2-3B-Instruct
Q4
2GB
111.43 tok/sEstimated
Auto-generated benchmark
google/gemma-2b
Q4
1GB
110.88 tok/sEstimated
Auto-generated benchmark
facebook/sam3
Q4
1GB
110.50 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen2.5-3B-Instruct
Q4
2GB
110.39 tok/sEstimated
Auto-generated benchmark
ibm-granite/granite-3.3-2b-instruct
Q4
1GB
109.60 tok/sEstimated
Auto-generated benchmark
nineninesix/kani-tts-2-en
Q4
1GB
108.96 tok/sEstimated
Auto-generated benchmark
unsloth/gemma-3-1b-it
Q4
1GB
108.79 tok/sEstimated
Auto-generated benchmark
allenai/OLMo-2-0425-1B
Q4
1GB
108.38 tok/sEstimated
Auto-generated benchmark
google/gemma-2-2b-it
Q4
1GB
106.14 tok/sEstimated
Auto-generated benchmark
TinyLlama/TinyLlama-1.1B-Chat-v1.0
Q4
1GB
105.12 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-3.2-1B
Q4
1GB
102.76 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen2.5-3B
Q4
2GB
102.56 tok/sEstimated
Auto-generated benchmark
google-bert/bert-base-uncased
Q4
1GB
102.04 tok/sEstimated
Auto-generated benchmark
deepseek-ai/deepseek-coder-1.3b-instruct
Q4
2GB
101.88 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-3.2-1B-Instruct
Q4
1GB
101.82 tok/sEstimated
Auto-generated benchmark
google-t5/t5-3b
Q4
2GB
101.16 tok/sEstimated
Auto-generated benchmark
bigcode/starcoder2-3b
Q4
2GB
100.91 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-3.2-3B
Q4
2GB
100.25 tok/sEstimated
Auto-generated benchmark
inference-net/Schematron-3B
Q4
2GB
100.22 tok/sEstimated
Auto-generated benchmark
google/embeddinggemma-300m
Q4
1GB
100.15 tok/sEstimated
Auto-generated benchmark
apple/OpenELM-1_1B-Instruct
Q4
1GB
99.78 tok/sEstimated
Auto-generated benchmark
google/gemma-3-1b-it
Q4
1GB
98.40 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-Guard-3-1B
Q4
1GB
97.95 tok/sEstimated
Auto-generated benchmark
context-labs/meta-llama-Llama-3.2-3B-Instruct-FP16
Q4
2GB
96.24 tok/sEstimated
Auto-generated benchmark
tencent/HunyuanOCR
Q4
1GB
95.78 tok/sEstimated
Auto-generated benchmark
meta-llama/Llama-2-7b-hf
Q4
4GB
95.62 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen2-0.5B-Instruct
Q4
3GB
95.62 tok/sEstimated
Auto-generated benchmark
microsoft/Phi-4-multimodal-instruct
Q4
4GB
95.54 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen3-8B-Base
Q4
4GB
95.47 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen3-Embedding-4B
Q4
2GB
95.44 tok/sEstimated
Auto-generated benchmark
tencent/HunyuanVideo-1.5
Q4
4GB
95.42 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen3-4B-Instruct-2507
Q4
2GB
95.38 tok/sEstimated
Auto-generated benchmark
Qwen/Qwen3-ASR-1.7B
Q4
2GB
95.34 tok/sEstimated
Auto-generated benchmark
microsoft/Phi-3.5-mini-instruct
Q4
2GB
95.29 tok/sEstimated
Auto-generated benchmark
ibm-granite/granite-3.3-8b-instruct
Q4
4GB
95.10 tok/sEstimated
Auto-generated benchmark
GSAI-ML/LLaDA-8B-Instruct
Q4
4GB
94.84 tok/sEstimated
Auto-generated benchmark
deepseek-ai/DeepSeek-R1-0528
Q4
4GB
94.80 tok/sEstimated
Auto-generated benchmark
zai-org/GLM-4.5-Air
Q4
4GB
94.53 tok/sEstimated
Auto-generated benchmark
microsoft/DialoGPT-small
Q4
4GB
94.52 tok/sEstimated
Auto-generated benchmark
unsloth/Llama-3.2-3B-Instruct
Q4
2GB
94.51 tok/sEstimated
Auto-generated benchmark
openai-community/gpt2
Q4
4GB
94.40 tok/sEstimated
Auto-generated benchmark
microsoft/DialoGPT-medium
Q4
4GB
94.35 tok/sEstimated
Auto-generated benchmark
liuhaotian/llava-v1.5-7b
Q4
4GB
94.34 tok/sEstimated
Auto-generated benchmark

Note: Performance estimates are calculated. Real results may vary. Methodology · Submit real data

Model compatibility

ModelQuantizationVerdictEstimated speedVRAM needed
openai-community/gpt2Q8Fits comfortably
56.75 tok/sEstimated
7GB (have 16GB)
openai-community/gpt2FP16Fits (tight)
30.16 tok/sEstimated
15GB (have 16GB)
Qwen/Qwen2.5-7B-InstructQ4Fits comfortably
87.55 tok/sEstimated
4GB (have 16GB)
Qwen/Qwen2.5-7B-InstructQ8Fits comfortably
66.20 tok/sEstimated
7GB (have 16GB)
Qwen/Qwen2.5-7B-InstructFP16Fits (tight)
30.53 tok/sEstimated
15GB (have 16GB)
Qwen/Qwen3-0.6BQ4Fits comfortably
83.05 tok/sEstimated
3GB (have 16GB)
Qwen/Qwen3-0.6BQ8Fits comfortably
56.38 tok/sEstimated
6GB (have 16GB)
Qwen/Qwen3-0.6BFP16Fits comfortably
33.45 tok/sEstimated
13GB (have 16GB)
Gensyn/Qwen2.5-0.5B-InstructQ4Fits comfortably
81.61 tok/sEstimated
3GB (have 16GB)
Gensyn/Qwen2.5-0.5B-InstructQ8Fits comfortably
66.55 tok/sEstimated
5GB (have 16GB)
Gensyn/Qwen2.5-0.5B-InstructFP16Fits comfortably
30.48 tok/sEstimated
11GB (have 16GB)
meta-llama/Llama-3.1-8B-InstructQ4Fits comfortably
78.90 tok/sEstimated
4GB (have 16GB)
meta-llama/Llama-3.1-8B-InstructQ8Fits comfortably
57.68 tok/sEstimated
9GB (have 16GB)
meta-llama/Llama-3.1-8B-InstructFP16Not supported
35.10 tok/sEstimated
17GB (have 16GB)
dphn/dolphin-2.9.1-yi-1.5-34bQ4Not supported
30.06 tok/sEstimated
17GB (have 16GB)
dphn/dolphin-2.9.1-yi-1.5-34bQ8Not supported
22.98 tok/sEstimated
35GB (have 16GB)
dphn/dolphin-2.9.1-yi-1.5-34bFP16Not supported
12.72 tok/sEstimated
70GB (have 16GB)
openai/gpt-oss-20bQ4Fits comfortably
48.00 tok/sEstimated
10GB (have 16GB)
openai/gpt-oss-20bQ8Not supported
34.70 tok/sEstimated
20GB (have 16GB)
openai/gpt-oss-20bFP16Not supported
19.86 tok/sEstimated
41GB (have 16GB)
google/gemma-3-1b-itQ4Fits comfortably
98.40 tok/sEstimated
1GB (have 16GB)
google/gemma-3-1b-itQ8Fits comfortably
66.15 tok/sEstimated
1GB (have 16GB)
google/gemma-3-1b-itFP16Fits comfortably
38.11 tok/sEstimated
2GB (have 16GB)
Qwen/Qwen3-Embedding-0.6BQ4Fits comfortably
93.29 tok/sEstimated
3GB (have 16GB)
Qwen/Qwen3-Embedding-0.6BQ8Fits comfortably
57.27 tok/sEstimated
6GB (have 16GB)
Qwen/Qwen3-Embedding-0.6BFP16Fits comfortably
30.64 tok/sEstimated
13GB (have 16GB)
Qwen/Qwen2.5-1.5B-InstructQ4Fits comfortably
87.55 tok/sEstimated
3GB (have 16GB)
Qwen/Qwen2.5-1.5B-InstructQ8Fits comfortably
58.64 tok/sEstimated
5GB (have 16GB)
Qwen/Qwen2.5-1.5B-InstructFP16Fits comfortably
33.65 tok/sEstimated
11GB (have 16GB)
facebook/opt-125mQ4Fits comfortably
85.88 tok/sEstimated
4GB (have 16GB)
facebook/opt-125mQ8Fits comfortably
63.41 tok/sEstimated
7GB (have 16GB)
facebook/opt-125mFP16Fits (tight)
35.45 tok/sEstimated
15GB (have 16GB)
TinyLlama/TinyLlama-1.1B-Chat-v1.0Q4Fits comfortably
105.12 tok/sEstimated
1GB (have 16GB)
TinyLlama/TinyLlama-1.1B-Chat-v1.0Q8Fits comfortably
80.34 tok/sEstimated
1GB (have 16GB)
TinyLlama/TinyLlama-1.1B-Chat-v1.0FP16Fits comfortably
37.51 tok/sEstimated
2GB (have 16GB)
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5Q4Fits comfortably
92.84 tok/sEstimated
4GB (have 16GB)
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5Q8Fits comfortably
61.20 tok/sEstimated
7GB (have 16GB)
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5FP16Fits (tight)
33.16 tok/sEstimated
15GB (have 16GB)
Qwen/Qwen3-4B-Instruct-2507Q4Fits comfortably
95.38 tok/sEstimated
2GB (have 16GB)
Qwen/Qwen3-4B-Instruct-2507Q8Fits comfortably
55.37 tok/sEstimated
4GB (have 16GB)
Qwen/Qwen3-4B-Instruct-2507FP16Fits comfortably
32.39 tok/sEstimated
9GB (have 16GB)
meta-llama/Llama-3.2-1B-InstructQ4Fits comfortably
101.82 tok/sEstimated
1GB (have 16GB)
meta-llama/Llama-3.2-1B-InstructQ8Fits comfortably
75.03 tok/sEstimated
1GB (have 16GB)
meta-llama/Llama-3.2-1B-InstructFP16Fits comfortably
36.37 tok/sEstimated
2GB (have 16GB)
openai/gpt-oss-120bQ4Not supported
16.60 tok/sEstimated
59GB (have 16GB)
openai/gpt-oss-120bQ8Not supported
10.98 tok/sEstimated
117GB (have 16GB)
openai/gpt-oss-120bFP16Not supported
7.06 tok/sEstimated
235GB (have 16GB)
Qwen/Qwen2.5-3B-InstructQ4Fits comfortably
110.39 tok/sEstimated
2GB (have 16GB)
Qwen/Qwen2.5-3B-InstructQ8Fits comfortably
70.25 tok/sEstimated
3GB (have 16GB)
openai-community/gpt2Q4Fits comfortably
94.40 tok/sEstimated
4GB (have 16GB)
openai-community/gpt2Q8
Fits comfortably7GB required · 16GB available
56.75 tok/sEstimated
openai-community/gpt2FP16
Fits (tight)15GB required · 16GB available
30.16 tok/sEstimated
Qwen/Qwen2.5-7B-InstructQ4
Fits comfortably4GB required · 16GB available
87.55 tok/sEstimated
Qwen/Qwen2.5-7B-InstructQ8
Fits comfortably7GB required · 16GB available
66.20 tok/sEstimated
Qwen/Qwen2.5-7B-InstructFP16
Fits (tight)15GB required · 16GB available
30.53 tok/sEstimated
Qwen/Qwen3-0.6BQ4
Fits comfortably3GB required · 16GB available
83.05 tok/sEstimated
Qwen/Qwen3-0.6BQ8
Fits comfortably6GB required · 16GB available
56.38 tok/sEstimated
Qwen/Qwen3-0.6BFP16
Fits comfortably13GB required · 16GB available
33.45 tok/sEstimated
Gensyn/Qwen2.5-0.5B-InstructQ4
Fits comfortably3GB required · 16GB available
81.61 tok/sEstimated
Gensyn/Qwen2.5-0.5B-InstructQ8
Fits comfortably5GB required · 16GB available
66.55 tok/sEstimated
Gensyn/Qwen2.5-0.5B-InstructFP16
Fits comfortably11GB required · 16GB available
30.48 tok/sEstimated
meta-llama/Llama-3.1-8B-InstructQ4
Fits comfortably4GB required · 16GB available
78.90 tok/sEstimated
meta-llama/Llama-3.1-8B-InstructQ8
Fits comfortably9GB required · 16GB available
57.68 tok/sEstimated
meta-llama/Llama-3.1-8B-InstructFP16
Not supported17GB required · 16GB available
35.10 tok/sEstimated
dphn/dolphin-2.9.1-yi-1.5-34bQ4
Not supported17GB required · 16GB available
30.06 tok/sEstimated
dphn/dolphin-2.9.1-yi-1.5-34bQ8
Not supported35GB required · 16GB available
22.98 tok/sEstimated
dphn/dolphin-2.9.1-yi-1.5-34bFP16
Not supported70GB required · 16GB available
12.72 tok/sEstimated
openai/gpt-oss-20bQ4
Fits comfortably10GB required · 16GB available
48.00 tok/sEstimated
openai/gpt-oss-20bQ8
Not supported20GB required · 16GB available
34.70 tok/sEstimated
openai/gpt-oss-20bFP16
Not supported41GB required · 16GB available
19.86 tok/sEstimated
google/gemma-3-1b-itQ4
Fits comfortably1GB required · 16GB available
98.40 tok/sEstimated
google/gemma-3-1b-itQ8
Fits comfortably1GB required · 16GB available
66.15 tok/sEstimated
google/gemma-3-1b-itFP16
Fits comfortably2GB required · 16GB available
38.11 tok/sEstimated
Qwen/Qwen3-Embedding-0.6BQ4
Fits comfortably3GB required · 16GB available
93.29 tok/sEstimated
Qwen/Qwen3-Embedding-0.6BQ8
Fits comfortably6GB required · 16GB available
57.27 tok/sEstimated
Qwen/Qwen3-Embedding-0.6BFP16
Fits comfortably13GB required · 16GB available
30.64 tok/sEstimated
Qwen/Qwen2.5-1.5B-InstructQ4
Fits comfortably3GB required · 16GB available
87.55 tok/sEstimated
Qwen/Qwen2.5-1.5B-InstructQ8
Fits comfortably5GB required · 16GB available
58.64 tok/sEstimated
Qwen/Qwen2.5-1.5B-InstructFP16
Fits comfortably11GB required · 16GB available
33.65 tok/sEstimated
facebook/opt-125mQ4
Fits comfortably4GB required · 16GB available
85.88 tok/sEstimated
facebook/opt-125mQ8
Fits comfortably7GB required · 16GB available
63.41 tok/sEstimated
facebook/opt-125mFP16
Fits (tight)15GB required · 16GB available
35.45 tok/sEstimated
TinyLlama/TinyLlama-1.1B-Chat-v1.0Q4
Fits comfortably1GB required · 16GB available
105.12 tok/sEstimated
TinyLlama/TinyLlama-1.1B-Chat-v1.0Q8
Fits comfortably1GB required · 16GB available
80.34 tok/sEstimated
TinyLlama/TinyLlama-1.1B-Chat-v1.0FP16
Fits comfortably2GB required · 16GB available
37.51 tok/sEstimated
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5Q4
Fits comfortably4GB required · 16GB available
92.84 tok/sEstimated
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5Q8
Fits comfortably7GB required · 16GB available
61.20 tok/sEstimated
trl-internal-testing/tiny-Qwen2ForCausalLM-2.5FP16
Fits (tight)15GB required · 16GB available
33.16 tok/sEstimated
Qwen/Qwen3-4B-Instruct-2507Q4
Fits comfortably2GB required · 16GB available
95.38 tok/sEstimated
Qwen/Qwen3-4B-Instruct-2507Q8
Fits comfortably4GB required · 16GB available
55.37 tok/sEstimated
Qwen/Qwen3-4B-Instruct-2507FP16
Fits comfortably9GB required · 16GB available
32.39 tok/sEstimated
meta-llama/Llama-3.2-1B-InstructQ4
Fits comfortably1GB required · 16GB available
101.82 tok/sEstimated
meta-llama/Llama-3.2-1B-InstructQ8
Fits comfortably1GB required · 16GB available
75.03 tok/sEstimated
meta-llama/Llama-3.2-1B-InstructFP16
Fits comfortably2GB required · 16GB available
36.37 tok/sEstimated
openai/gpt-oss-120bQ4
Not supported59GB required · 16GB available
16.60 tok/sEstimated
openai/gpt-oss-120bQ8
Not supported117GB required · 16GB available
10.98 tok/sEstimated
openai/gpt-oss-120bFP16
Not supported235GB required · 16GB available
7.06 tok/sEstimated
Qwen/Qwen2.5-3B-InstructQ4
Fits comfortably2GB required · 16GB available
110.39 tok/sEstimated
Qwen/Qwen2.5-3B-InstructQ8
Fits comfortably3GB required · 16GB available
70.25 tok/sEstimated
openai-community/gpt2Q4
Fits comfortably4GB required · 16GB available
94.40 tok/sEstimated

Note: Performance estimates are calculated. Real results may vary. Methodology · Submit real data

Alternative GPUs

RTX 5070
12GB

Explore how RTX 5070 stacks up for local inference workloads.

RTX 4060 Ti 16GB
16GB

Explore how RTX 4060 Ti 16GB stacks up for local inference workloads.

RX 6800 XT
16GB

Explore how RX 6800 XT stacks up for local inference workloads.

RTX 4070 Super
12GB

Explore how RTX 4070 Super stacks up for local inference workloads.

RTX 3080
10GB

Explore how RTX 3080 stacks up for local inference workloads.

Can it play popular games?

Cyberpunk 2077
8GB VRAM

RPG • 2020

Baldur's Gate 3
8GB VRAM

RPG • 2023

Hogwarts Legacy
12GB VRAM

Action RPG • 2023

Starfield
8GB VRAM

RPG • 2023

Alan Wake 2
12GB VRAM

Survival Horror • 2023

Elden Ring
8GB VRAM

Action RPG • 2022

Black Myth: Wukong
12GB VRAM

Action RPG • 2024

Grand Theft Auto VI
12GB VRAM

Action Adventure • 2025

Resident Evil 4 Remake
12GB VRAM

Survival Horror • 2023

Marvel's Spider-Man Remastered
12GB VRAM

Action • 2022

The Last of Us Part I
12GB VRAM

Action Adventure • 2023

Red Dead Redemption 2
8GB VRAM

Action Adventure • 2019

View all 64 compatible games