FreeLLMAll models › Compare
Gemma 4 31B vs nvidia-nemotron-3-ultra-550b-a55b-nvfp4
Gemma 4 31B and nvidia-nemotron-3-ultra-550b-a55b-nvfp4 are about even on Arena (1443 vs 1445): in a blind vote each wins about half the time.
Gemma 4 31B1443Arena chat · #67 of 416
nvidia-nemotron-3-ultra-550b-a55b-nvfp41445Arena chat · #64 of 416
Arena scores by category
| Category | Gemma 4 31B | nvidia-nemotron-3-ultra-550b-a55b-nvfp4 | Ahead |
|---|---|---|---|
| Chat | 1443 | 1445 | about even |
Scores from blind human votes on Arena. A gap under 10 points is too small to call.
Size and where to use them free
| Gemma 4 31B | nvidia-nemotron-3-ultra-550b-a55b-nvfp4 | |
|---|---|---|
| Made by | nvidia | |
| Size | 31.3 B | 560.5 B (55 B active) |
| 4-bit download | 20 GB | 341 GB |
| Use it free | free API on OpenRouter · download and run at home · Apache 2.0 license | download and run at home · OpenMDW-1.1 license |
Speed at home
| Hardware | Gemma 4 31B | nvidia-nemotron-3-ultra-550b-a55b-nvfp4 |
|---|---|---|
| RTX 3090 | 33 tok/s | does not fit |
| RTX 4090 | 36 tok/s | does not fit |
| RTX 5090 | 62 tok/s | does not fit |
| RX 7900 XTX | 34 tok/s | does not fit |
| Mac M4 Max 128 GB | 19 tok/s | does not fit |
| Ryzen AI Max+ 395 128 GB | 9.4 tok/s | does not fit |
Estimates at 4-bit with an 8K context, 64 GB of system RAM for offload, calibrated on 44 real llama.cpp tests.
Try them yourself
Ask both the same question side by side in Try & compare, or vote blind between free models in Free Arena.
More comparisons
- Gemma 4 31B vs MiMo-V2.6-Pro
- Gemma 4 31B vs Kimi K3 (Max)
- Gemma 4 31B vs GLM-5.3 (Max)
- Gemma 4 31B vs GLM-5.3-Flash
- Gemma 4 31B vs GLM-5.2 (Max)
- Gemma 4 31B vs MiMo-V2.5-Pro
- nvidia-nemotron-3-ultra-550b-a55b-nvfp4 vs MiMo-V2.6-Pro
- nvidia-nemotron-3-ultra-550b-a55b-nvfp4 vs Kimi K3 (Max)
- nvidia-nemotron-3-ultra-550b-a55b-nvfp4 vs GLM-5.3 (Max)
- nvidia-nemotron-3-ultra-550b-a55b-nvfp4 vs GLM-5.3-Flash