Can SmolLM2 1.7B run on the RX 7900 GRE?
Yes. At 8K context the best fit is F16, using about 5.2 GB of the 15.2 GB usable on the RX 7900 GRE. Estimated speed ceiling: ~168 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | Max tok/s (est., 8K) |
|---|---|---|---|
| 2-bit (Q2_K) | 1.8 GB Fits | 2.6 GB Fits | 898 |
| 3-bit (Q3_K_M) | 2.0 GB Fits | 2.8 GB Fits | 689 |
| 4-bit (Q4_K_M) • | 2.3 GB Fits | 3.0 GB Fits | 533 |
| 5-bit (Q5_K_M) | 2.4 GB Fits | 3.1 GB Fits | 472 |
| 6-bit (Q6_K) | 2.6 GB Fits | 3.3 GB Fits | 410 |
| 8-bit (Q8_0) • | 3.0 GB Fits | 3.7 GB Fits | 311 |
| 16-bit (F16/BF16) | 4.4 GB Fits | 5.2 GB Fits | 168 |
• real GGUF file size from namindu94/SinhalaLLM-1.7B. Speed is a bandwidth ceiling (576 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs SmolLM2 1.7B: Arc B580 (12 GB, launch MSRP $249) at F16, fully in VRAM at 8K context.
- Largest popular model that fits the RX 7900 GRE: gpt-oss 20B (21B, MoE) at Q8_0, about 12.3 GB at 8K context.
"Runs" and "fits" mean the whole model in GPU memory at 8K context (4-bit or better for the model pick). Prices are US launch MSRPs, not today's street prices. "Popular" means the 60 most-downloaded models we track.
More
SmolLM2 1.7B on every GPU and Mac · Every model that runs on the RX 7900 GRE · Other models that fit 16 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (HuggingFaceTB/SmolLM2-1.7B), AMD specs. Model data last checked 2026-11-17; refreshed hourly.