Can Laptop · Ryzen AI 9 HX 370 (Radeon 890M) · 32 GB run DeepSeek-R1-0528-Qwen3-8B?
32 GB unified · 120 GB/s memory bandwidth · DeepSeek-R1-0528-Qwen3-8B has 8.2B parameters.
Short answer
Yes, comfortable for chat; borderline for agentic coding.
Best quantization for chat: Q5_K_M (5.6 GB of weights) on Lemonade. Generation: 10–16 tok/s, leaving about 16 GB free for your other apps.
How it rates for each workload
Same model and machine, different jobs. Agents and long documents send far bigger prompts than chat, so they need much more speed and memory.
| Workload | Verdict | Quant | Generation | First reply | Context | |
|---|---|---|---|---|---|---|
| Chatwith Open WebUI | Comfortable | Q5_K_M | 10–16 tok/s | ≈3 s | 8K | Details |
| Coding questionswith Continue | Acceptable | Q6_K | 8–13 tok/s | ≈20 s | 16K | Details |
| Coding in a repositorywith Aider | Acceptable | Q5_K_M | 7–12 tok/s | ≈65 s | 32K | Details |
| Agentic codingwith OpenCode | Borderline | Q6_K | 6–10 tok/s | ≈90 s | 32K | Details |
| Long documents (64K)with Open WebUI | Borderline | Q6_K | 4.2–7 tok/s | ≈5 min | 64K | Details |
Speeds are bandwidth-based estimates; no direct benchmark exists for this pair. “First reply” is the time to read the workload’s opening prompt. How we calculate
Get it running
Copy-paste commands for the Q5_K_M build with a 32K context. Ollama and llama.cpp download the file straight from Hugging Face.
ollama run hf.co/unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF:Q5_K_M
Downloads 5.85 GB (Q5_K_M). Ollama's default context is small: set OLLAMA_CONTEXT_LENGTH=32768 before `ollama serve` for long prompts.
Other models for Laptop · Ryzen AI 9 HX 370 (Radeon 890M) · 32 GB
The best-rated models for chat on this machine.
- LFM2.5 24B-A2B35–60 tok/sExcellent
- gpt-oss-20b20–35 tok/sExcellent
- MiniCPM5 2B20–35 tok/sExcellent
- Qwen3.5 4B15–25 tok/sExcellent
- Llama 3.2 3B Instruct20–35 tok/sExcellent
- Gemma 4 26B-A4B16–25 tok/sComfortable
Cheapest machines for DeepSeek-R1-0528-Qwen3-8B
Hardware rated comfortable or better for chat, lowest price first.
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.