DeepSeek-V4-Flash vs LFM2-24B-A2B
DeepSeek vs Liquid AI — benchmarks, pricing, and capabilities side by side.
- •DeepSeek-V4-Flash has the higher intelligence index (89.4 vs 47.4)
- •LFM2-24B-A2B is cheaper ($0.03 vs $0.10 per 1M input)
- •LFM2-24B-A2B is faster
- •DeepSeek-V4-Flash has a larger context window (1M)
| DeepSeek-V4-Flash | LFM2-24B-A2B | |
|---|---|---|
| Intelligence index | 89.4 | 47.4 |
| Developer | DeepSeek | Liquid AI |
| Type | LLM | LLM |
| Access | Open weights | Open weights |
| Context window | 1,048,576 tokens | 128,000 tokens |
| Input price | $0.10 / 1M | $0.03 / 1M |
| Output price | $0.20 / 1M | $0.12 / 1M |
| Speed | 109 tok/s | 208 tok/s |
| Released | April 24, 2026 | February 25, 2026 |
| Parameters | 284B (13B active) | — |
| Input modalities | Text | Text |
| Output modalities | Text | Text |
Shared benchmarks
DeepSeek-V4-Flash
LFM2-24B-A2B
GPQA Diamond
89.4
47.4
Humanity’s Last Exam
32.1
4.4