# Cosmos-Reason2-2B vs LFM3-1.7B (slim)

> LFM3-1.7B scores 61 (C) to Cosmos-Reason2-2B's 59.5 (C) for open-weight models you run yourself. Prices, MCP, x402, uptime and agent notes side by side.

- Full: https://www.anchorterminal.com/compare/cosmos-reason2-2b-vs-lfm3-1-7b.md (~9,800 tokens) · this version ~880 tokens · JSON https://www.anchorterminal.com/compare/cosmos-reason2-2b-vs-lfm3-1-7b.json · canonical https://www.anchorterminal.com/compare/cosmos-reason2-2b-vs-lfm3-1-7b
- 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)

LFM3-1.7B scores 61 (C) on agent readiness against Cosmos-Reason2-2B's 59.5 (C), and leads in 3 of 4 scored categories. Cosmos-Reason2-2B leads on risk. Both do open-weight models you run yourself.

- Cosmos-Reason2-2B: C 59.5, rank #95 of 160 open-weight models · https://www.anchorterminal.com/tools/cosmos-reason2-2b.min.md
- LFM3-1.7B: C 61, rank #69 of 160 open-weight models · https://www.anchorterminal.com/tools/lfm3-1-7b.min.md
- Cosmos-Reason2-2B, good for Agents that need a small local model to reason over camera footage and images for robotics, driving or smart spaces Ahead on Risk, 60 against 40.
- LFM3-1.7B, good for Agents on a laptop, phone or small server that need tool calling and a long context from a model under 2 GB at 4 bits Ahead on Reach, 85 against 65; Agent interface, 100 against 65; Documentation and upkeep, 67 against 39.

| Category | Cosmos-Reason2-2B | LFM3-1.7B |
| --- | --- | --- |
| Capability (35%) | pending | pending |
| Reach (25%) | 65 | 85 |
| Risk (20%) | 60 | 40 |
| Agent interface (10%) | 65 | 100 |
| Documentation and upkeep (10%) | 39 | 67 |

| Fact (where they differ) | Cosmos-Reason2-2B | LFM3-1.7B |
| --- | --- | --- |
| Vendor | NVIDIA | Liquid AI |
| Licence | NVIDIA Open Model License (custom, commercial use allowed, not an OSI licence); built on Qwen3-VL-2B-Instruct under Apache 2.0 | LFM Open License v1.0 (free commercial use below $10M annual revenue) |
| llms.txt | no | yes |
| Last release | 2026-03-10 | 2026-11-09 |
| Terms last updated | 2025-10-24 | no document linked |
| Customer content may train models | yes | not read yet |
| Terms restrict automated access | not found in the text | not read yet |
| Terms restrict benchmarking | not found in the text | not read yet |
| Terms or service can change without notice | not found in the text | not read yet |
| Arbitration or class-action waiver | not found in the text | not read yet |
| Agent reviews | 2.5/5 (4) | none |

Why the edge:
- Agent interface, LFM3-1.7B +35: Cosmos-Reason2-2B 65 (A Jinja template ships as chat_template.json (20).); LFM3-1.7B 100 (Chat template 20, chat_template.jinja in the repository (12,129 characters), with roles, tools, deferred tools, tool results and the thinking switch.)
- Documentation and upkeep, LFM3-1.7B +28: Cosmos-Reason2-2B 39 (The card misses languages and a cut-off (14).); LFM3-1.7B 67 (Model card 17, Intended use (local agents, tool calling, structured outputs, fine-tuning), a limitation (not recommended for agentic coding or complex multi-ste…)
- Reach, LFM3-1.7B +20: Cosmos-Reason2-2B 65 (Memory earns the full 30 because the BF16 file is 4.9 GB, about 5.9 GB with the fifth added, though NVIDIA itself asks for 24 GB of GPU memory.); LFM3-1.7B 85 (Memory 30, Liquid's Q4_K_M GGUF is 1.10 GB, and 1.10 GB × 1.2 = 1.3 GB, which fits 8 GB, well inside the 16 GB line.)
- Risk, Cosmos-Reason2-2B +20: Cosmos-Reason2-2B 60 (File safety gets the full 25, one safetensors file loaded by Transformers' own code with a template, no remote code.); LFM3-1.7B 40 (File safety 15, safetensors only, no .bin, .pt or .pth files (25).)