RunsOnMy ModelsGPUs & MacsBy VRAMBest GPUCan I run…?API pricesVRAM calculatorNew

Can ERNIE 4.5 21B A3B Thinking run on the RTX 4090?

Yes. At 8K context the best fit is Q6_K, using about 17.9 GB of the 22.8 GB usable on the RTX 4090. Estimated speed ceiling: ~56 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctx32K ctx128K ctxMax tok/s (est., 8K)
2-bit (Q2_K) •8.4 GB Fits8.6 GB Fits9.9 GB Fits15.1 GB Fits125
3-bit (Q3_K_M) •10.4 GB Fits10.6 GB Fits11.9 GB Fits17.2 GB Fits98
4-bit (Q4_K_M) •13.4 GB Fits13.6 GB Fits14.9 GB Fits20.2 GB Fits75
5-bit (Q5_K_M) •15.5 GB Fits15.7 GB Fits17.0 GB Fits22.3 GB Tight64
6-bit (Q6_K) •17.7 GB Fits17.9 GB Fits19.2 GB Fits24.4 GB No56
8-bit (Q8_0) •22.4 GB Tight22.6 GB Tight24.0 GB No29.2 GB No43
16-bit (F16/BF16) •41.5 GB No41.7 GB No43.0 GB No48.3 GB No—

• real GGUF file size from bartowski/baidu_ERNIE-4.5-21B-A3B-Thinking-GGUF. Speed is a bandwidth ceiling (1008 GB/s), not a benchmark. Methodology.

What fits instead

"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

ERNIE 4.5 21B A3B Thinking on every GPU and Mac · Every model that runs on the RTX 4090 · 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 (baidu/ERNIE-4.5-21B-A3B-Thinking), NVIDIA specs. Model data last checked 2026-11-17; refreshed hourly.