Model comparison
Gemma 4 12B and Nemotron 3 Ultra compared on price, context window, capabilities and published benchmarks. Data comes from the OpenModels registry.
| Specification | Gemma 4 12B | Nemotron 3 Ultra |
|---|---|---|
| Developer | Nvidia | |
| Released | Jun 4, 2026 | Jun 4, 2026 |
| Context window | 262K tokens | 1.05M tokens |
| Input price / 1M | $0.10via Hugging Face Inference | $0.40via Deep Infra |
| Output price / 1M | $0.10via Hugging Face Inference | $1.60via Deep Infra |
| Providers | 3 | 4 |
| License | apache-2.0 | other |
| Input modalities | audio, code, image, text, video | code, text |
| Capabilities | chat, completion, vision, audio, code-generation, reasoning, function-calling | chat, completion, function-calling, code-generation, reasoning |
| Reasoning | Yes | Yes |
| Tool calling | Yes | Yes |
No published benchmark scores recorded for these models yet.