LFM2 is a new generation of hybrid models developed by Liquid AI, specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency.
Model details →LiquidAI: LFM2-2.6B vs Qwen: Qwen3 30B A3B Thinking 2507
Qwen: Qwen3 30B A3B Thinking 2507 wins on 3 of 6 axes — pricing and capability skew in its favour for most workloads.
Qwen3-30B-A3B-Thinking-2507 is a 30B parameter Mixture-of-Experts reasoning model optimized for complex tasks requiring extended multi-step thinking. The model is designed specifically for “thinking mode,” where internal reasoning traces ar…
Model details →Side-by-side comparison
| Capability | LiquidAI: LFM2-2.6B | Qwen: Qwen3 30B A3B Thinking 2507 | Winner |
|---|---|---|---|
| Context window Maximum number of input tokens the model can attend to in a single request. | 33K | 82K | 🏆 Qwen: Qwen3 30B A3B Thinking 2507 |
| Input price (per 1M) Cost per million input tokens billed by the provider. | $0.01 | $0.20 | 🏆 LiquidAI: LFM2-2.6B |
| Output price (per 1M) Cost per million output tokens billed by the provider. | $0.02 | $2.40 | 🏆 LiquidAI: LFM2-2.6B |
| Tool / function calling First-class support for emitting structured tool calls. | — | Yes | 🏆 Qwen: Qwen3 30B A3B Thinking 2507 |
| Vision input Accepts image inputs alongside text. | — | — | Tie |
| Reasoning mode Internal chain-of-thought / extended-thinking support. | — | Yes | 🏆 Qwen: Qwen3 30B A3B Thinking 2507 |
Frequently asked questions
Is LiquidAI: LFM2-2.6B better than Qwen: Qwen3 30B A3B Thinking 2507?
Qwen: Qwen3 30B A3B Thinking 2507 wins on 3 of 6 axes — pricing and capability skew in its favour for most workloads.
What's the price difference between LiquidAI: LFM2-2.6B and Qwen: Qwen3 30B A3B Thinking 2507?
Input: $0.01 vs $0.20 per 1M tokens. Output: $0.02 vs $2.40 per 1M tokens.
What context windows do LiquidAI: LFM2-2.6B and Qwen: Qwen3 30B A3B Thinking 2507 support?
LiquidAI: LFM2-2.6B supports up to 33K tokens. Qwen: Qwen3 30B A3B Thinking 2507 supports up to 82K tokens.
Do both LiquidAI: LFM2-2.6B and Qwen: Qwen3 30B A3B Thinking 2507 support tool calling?
LiquidAI: LFM2-2.6B: not advertised. Qwen: Qwen3 30B A3B Thinking 2507: yes.
Compare LiquidAI: LFM2-2.6B with other models
vs Anthropic Claude Haiku Latestvs Anthropic Claude Sonnet Latestvs Anthropic: Claude Fable Latestvs Anthropic: Claude Opus Latest