Can Falcon H1R 7B run on the RX 7900 XT?
Yes. At 8K context the best fit is F16, using about 15.1 GB of the 19.0 GB usable on the RX 7900 XT. Estimated speed ceiling: ~53 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 3.5 GB Fits | 3.7 GB Fits | 4.7 GB Fits | 8.8 GB Fits | 276 |
| 3-bit (Q3_K_M) • | 4.3 GB Fits | 4.4 GB Fits | 5.5 GB Fits | 9.6 GB Fits | 217 |
| 4-bit (Q4_K_M) • | 5.1 GB Fits | 5.3 GB Fits | 6.3 GB Fits | 10.4 GB Fits | 174 |
| 5-bit (Q5_K_M) • | 5.8 GB Fits | 6.0 GB Fits | 7.0 GB Fits | 11.2 GB Fits | 148 |
| 6-bit (Q6_K) • | 6.6 GB Fits | 6.8 GB Fits | 7.8 GB Fits | 12.0 GB Fits | 128 |
| 8-bit (Q8_0) • | 8.3 GB Fits | 8.5 GB Fits | 9.5 GB Fits | 13.7 GB Fits | 99 |
| 16-bit (F16/BF16) • | 15.0 GB Fits | 15.1 GB Fits | 16.2 GB Fits | 20.3 GB No | 53 |
• real GGUF file size from unsloth/Falcon-H1R-7B-GGUF. Speed is a bandwidth ceiling (800 GB/s), not a benchmark. Methodology.
What fits instead
- Cheapest listed GPU that runs Falcon H1R 7B: Arc B580 (12 GB, launch MSRP $249) at Q8_0, fully in VRAM at 8K context.
- Largest popular model that fits the RX 7900 XT: Qwen3.8 27B (28B) at Q4_K_M, about 16.8 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
Falcon H1R 7B on every GPU and Mac · Every model that runs on the RX 7900 XT · Other models that fit 24 GB VRAM · Try other settings in the calculator
Guides: Best GPU for local LLMs by budget · Can I run gpt-oss, Llama 70B or Qwen3? · Mac vs PC for local LLMs
Data: Hugging Face Hub (tiiuae/Falcon-H1R-7B), AMD specs. Model data last checked 2026-11-17; refreshed hourly.