$ diff lfm2-5-24b-a2b deepseek-v4-1-flash
LFM2.5-24B-A2B vs DeepSeek V4.1-Flash
Values come from LFM2.5-24B-A2B and DeepSeek V4.1-Flash, where each is cited to its source. This page states no benchmark result and ranks neither model — it puts two published specifications next to each other. Where a lab has not published a figure, the row says so rather than guessing.
Liquid AILFM2.5-24B-A2B
- context
- 131K
- weights
- —
- $/M in
- $0.1
- $/M out
- $1
- context
- 1M
- weights
- open
- $/M in
- $0.15
- $/M out
- $0.6
What actually differs
- Context window
- DeepSeek V4.1-Flash takes 1M against 131K — 7.6× more room in a single request.
- Input price
- LFM2.5-24B-A2B at $0.1/M against $0.15/M — 1.5× cheaper to feed. Standard rates: one of these labs also quotes a lower cached-input tier, which applies only when a prefix is reused — see the full spec below.
- Output price
- DeepSeek V4.1-Flash at $0.6/M against $1/M — 1.7× cheaper to generate. Output dominates the bill on most agentic workloads, where the model writes far more than it reads.
- Recency
- LFM2.5-24B-A2B shipped 33 days after DeepSeek V4.1-Flash (2026-10-13 vs 2026-09-10).
Full spec
| Attribute | LFM2.5-24B-A2B | DeepSeek V4.1-Flash |
|---|---|---|
| Developer | Liquid AI | DeepSeek |
| Released | 2026-10-13 | 2026-09-10 |
| Context window | 131,072 tokens | 1,000,000 |
| Pricing | $0.10/M input · $0.05/M cached input · $1.00/M output | $0.15/M input (cache miss) · $0.003/M input (cache hit) · $0.60/M output, off-peak; peak rates double — see Pricing |
| License | not recorded | MIT (open-weight) |
| Availability | Downloadable base and post-trained weights; Liquid Playground and API | DeepSeek API, Hugging Face (`deepseek-ai/DeepSeek-V4.1-Flash`) |