RX 7900 XT 20GBBudget pick
20 GB VRAM (+4)800 GB/s (+240)
C
Raises estimated decode speed by about 91%.98.4 tok/s decode
Raises estimated decode speed by about 91%.
Adds memory headroom for longer context windows and future model growth.
~$899 MSRP
Can DeepSeek R1 0528 Qwen3 8B run on Intel Arc A770 16GB?
YES — Runs Great
C51Usable
○Estimated from fit model
DeepSeek R1 0528 Qwen3 8B needs ~8.3 GB VRAM. Intel Arc A770 16GB has 16.0 GB. With Q4_K_M quantization, expect ~52 tok/s.
Runtime: llama.cppCapacity: RoomyBandwidth: MediumStack: StandardBottleneck: Balanced
Share:
Operating mode
Choose the run profile you care about
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
Q4_K_M (Medium quality) — 8.3 GB, 51.6 tok/s, Runs well
8.3 GB required16.0 GB available
52% VRAM used
Fit status
Runs well
Decode
51.6 tok/s
TTFT
3749 ms
Safe context
147K
Memory
8.3 GB / 16.0 GB
Memory breakdown
Weights4.9 GB
KV Cache0.9 GB
Runtime0.9 GB
Headroom1.6 GB
See how fast it feels
See how fast it feelsDeepSeek R1 0528 Qwen3 8B on Intel Arc A770 16GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step.
Given information:
- Speed = 60 mph (miles per hour)
- Time = 2.5 hours
Formula: Distance = Speed x Time
Calculation:
Distance = 60 mph x 2.5 h
= 150 miles
The train travels 150 miles.
Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time.
Breakdown:
- First half: 1.25 hours at 60 mph = 75 miles
- Stop: 20 minutes (no distance)
- Second half: 1.25 hours at 60 mph = 75 miles
Total distance: 150 miles (same)
Total time: 2.5 h + 0.33 h = 2.83 hours
Average speed: 150 / 2.83 = 53 mph
The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 51.6 tok/s decode · 3.7s TTFT (warm) · 129 tok/s prefill
What limits this setup
The raw memory story may look fine, but the software ecosystem is still a constraint here.
Runtime ecosystem is narrower than CUDA
Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.
Best improvement path
Prefer CUDA if you want the path of least resistance
If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | C | Runs well | 51.6 tok/s | 2045 ms | 147K |
| Coding | C | Runs well | 51.6 tok/s | 3749 ms | 147K |
| Agentic Coding | C | Runs well | 51.6 tok/s | 5453 ms | 147K |
| Reasoning | C | Runs well | 51.6 tok/s | 4431 ms | 147K |
| RAG | C | Runs well | 51.6 tok/s | 6817 ms | 147K |
Quantization options
How DeepSeek R1 0528 Qwen3 8B (8B params) fits at each quantization level on Intel Arc A770 16GB (16.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.1 GB | Low | C47 |
Q3_K_S | 3 | 3.9 GB | Low | C48 |
NVFP4 | 4 | 4.5 GB | Medium | C48 |
Q4_K_M | 4 | 4.9 GB | Medium | C49 |
Q5_K_M | 5 | 5.8 GB | High | C50 |
Q6_K | 6 | 6.6 GB | High | C50 |
Q8_0Best for your GPU | 8 | 8.6 GB | Very High | C51 |
F16 | 16 | 16.4 GB | Maximum | F0 |
Get started
Copy-paste commands to run DeepSeek R1 0528 Qwen3 8B on your machine.
Run
lms load hf-maziyarpanahi--deepseek-r1-0528-qwen3-8b-gguf && lms server startUpgrade options
Hardware that runs DeepSeek R1 0528 Qwen3 8B well
20 GB VRAM (+4)640 GB/s (+80)
C
Raises estimated decode speed by about 98%.102.3 tok/s decode
Raises estimated decode speed by about 98%.
Adds memory headroom for longer context windows and future model growth.
This is not only a hardware jump. It also gives you a cleaner runtime ecosystem for local LLM tooling.
~$2,000 MSRP