$ diff kolibri-1 kimi-k2-8-preview
Kolibri-1 vs Kimi K2.8 Preview
Values come from Kolibri-1 and Kimi K2.8 Preview, 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.
Aleph AlphaKolibri-1
- context
- 16K
- weights
- open
- $/M in
- —
- $/M out
- —
- context
- 1M
- weights
- closed
- $/M in
- —
- $/M out
- —
What actually differs
- Context window
- Kimi K2.8 Preview takes 1M against 16K — 61.0× more room in a single request.
- Weights
- Kolibri-1 publishes weights (Apache 2.0); Kimi K2.8 Preview is API-only. That decides self-hosting, air-gapped deployment and fine-tuning before any capability question does.
- Recency
- Kolibri-1 shipped 22 days after Kimi K2.8 Preview (2026-10-03 vs 2026-09-11).
Full spec
| Attribute | Kolibri-1 | Kimi K2.8 Preview |
|---|---|---|
| Developer | Aleph Alpha | Moonshot AI |
| Released | 2026-10-03 | 2026-09-11 |
| Context window | 16,384 tokens native; up to 1,048,576 extended | 1,000,000 |
| Pricing | not recorded | not recorded |
| License | Apache 2.0 | proprietary |
| Availability | full weights on Hugging Face | Kimi API platform, Kimi Code, Kimi Work |