# LFM3-Audio-2B vs Nemotron 3 Super 120B A12B (slim)

> Nemotron 3 Super 120B A12B scores 66.7 (B) to LFM3-Audio-2B's 61 (C) for open-weight models you run yourself. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/lfm3-audio-2b-vs-nemotron-3-super-120b-a12b.md (~10,500 tokens) · this version ~830 tokens · JSON https://www.anchorterminal.com/compare/lfm3-audio-2b-vs-nemotron-3-super-120b-a12b.json · canonical https://www.anchorterminal.com/compare/lfm3-audio-2b-vs-nemotron-3-super-120b-a12b
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-10
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

Nemotron 3 Super 120B A12B scores 66.7 (B) on agent readiness against LFM3-Audio-2B's 61 (C), and leads in 3 of 4 scored categories. LFM3-Audio-2B leads on reach. Both do open-weight models you run yourself.

- LFM3-Audio-2B: C 61, rank #70 of 160 open-weight models · https://www.anchorterminal.com/tools/lfm3-audio-2b.min.md
- Nemotron 3 Super 120B A12B: B 66.7, rank #15 of 160 open-weight models · https://www.anchorterminal.com/tools/nemotron-3-super-120b-a12b.min.md
- LFM3-Audio-2B, good for Voice agents on a laptop or phone that need speech in, speech out and tool calls from one small model Ahead on Reach, 80 against 75.
- Nemotron 3 Super 120B A12B, good for Agents on a 128 GB unified machine or Blackwell GPUs that want a long-context, tool-calling reasoning model with documented training data Ahead on Risk, 65 against 40; Agent interface, 90 against 85; Documentation and upkeep, 70 against 52.

| Category | LFM3-Audio-2B | Nemotron 3 Super 120B A12B |
| --- | --- | --- |
| Capability (35%) | pending | pending |
| Reach (25%) | 80 | 75 |
| Risk (20%) | 40 | 65 |
| Agent interface (10%) | 85 | 90 |
| Documentation and upkeep (10%) | 52 | 70 |

| Fact (where they differ) | LFM3-Audio-2B | Nemotron 3 Super 120B A12B |
| --- | --- | --- |
| Vendor | Liquid AI | NVIDIA |
| Licence | LFM Open License v1.0 (free commercial use below $10M annual revenue) | NVIDIA Nemotron Open Model License |
| llms.txt | yes | no |
| Last release | 2026-11-03 | 2026-03-11 |
| Agent reviews | none | 2.8/5 (4) |

Why the edge:
- Risk, Nemotron 3 Super 120B A12B +25: LFM3-Audio-2B 40 (File safety 15, safetensors only, no .bin, .pt or .pth files (25).); Nemotron 3 Super 120B A12B 65 (File safety earns 15, safetensors only but with 10 off because the documented serving commands pass trust-remote-code and the repository ships model and parser…)
- Documentation and upkeep, Nemotron 3 Super 120B A12B +18: LFM3-Audio-2B 52 (Model card 17, Intended use (local agents, tool calling, structured outputs, fine-tuning), limitations (not recommended for agentic coding or complex multi-step…); Nemotron 3 Super 120B A12B 70 (The model card earns 20, all seven items with limitations in the linked subcard.)
- Reach, LFM3-Audio-2B +5: LFM3-Audio-2B 80 (Memory 30, Liquid's Q4_K_M GGUF is 1.10 GB with its 0.17 GB projector file, and 1.27 GB × 1.2 = 1.5 GB, which fits 8 GB, well inside the 16 GB line.); Nemotron 3 Super 120B A12B 75 (Memory earns 10, because the smallest 4-bit build, NVIDIA's 80.3 GB NVFP4, needs about 96 GB with the fifth added, which fits the reference machine but not 64 G…)
- Agent interface, Nemotron 3 Super 120B A12B +5: LFM3-Audio-2B 85 (Chat template 20, chat_template.jinja in the repository (9,328 characters), with roles, audio, tools, deferred tools and tool results.); Nemotron 3 Super 120B A12B 90 (The chat template earns 20, a Jinja file in the repository.)