Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Kanana: Compute-efficient Bilingual Language Models
| Date | Stars |
|---|---|
| 2026-07-31 | 285 |
| 2026-08-06 | 285 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<p align="center">
<br>
<picture>
<source media="(prefers-color-scheme: dark)" srcset="assets/logo/kanana-logo-light.png">
<source media="(prefers-color-scheme: light)" srcset="assets/logo/kanana-logo-dark.png">
<img alt="Kanana Logo" src="assets/logo/kanana-logo-light.png" width="400" style="margin: 40px auto;">
</picture>
</br>
<p align="center">
🤗 <a href="https://kko.kakao.com/kananallm">Kanana-1.5 HF Models</a>   |
  📕 <a href="https://tech.kakao.com/search?q=kanana">Blogs</a>  
<br>
## News 🔥
- ✨`2025/07/24`: Published a [blog post](https://tech.kakao.com/posts/716) about `Kanana-1.5-15.7B-A3B` models and released 🤗[HF model weights](https://kko.kakao.com/kananallm).
- 📕`2025/05/23`: Published a [blog post](https://tech.kakao.com/posts/707) about `Kanana 1.5` models and released 🤗[HF model weights](https://kko.kakao.com/kananallm).
- 📜`2025/02/27`: Released [Technical Report](https://arxiv.org/abs/2502.18934) and 🤗[HF model weights](https://huggingface.co/collections/kakaocorp/kanana-nano-21b-67a326cda1c449c8d4172259).
- 📕`2025/01/10`: Published a [blog post](https://tech.kakao.com/posts/682) about the development of `Kanana Nano` model.
- 📕`2024/11/14`: Published blog posts ([pre-training](https://tech.kakao.com/posts/661), [post-training](https://tech.kakao.com/posts/662)) about the development of `Kanana` models.
- ▶️`2024/11/06`: Published a [presentation video](https://youtu.be/HTBl142x9GI?si=o_we6t9suYK8DfX3) about the development of the `Kanana` models.
<br>
## Table of Contents
- [Kanana 1.5](#kanana-15)
- [Performance](#performance)
- [Base Model Evaluation](#base-model-evaluation)
- [Instruct Model Evaluation](#instruct-model-evaluation)
- [Long Context](#long-context)
- [Processing 32K+ Length](#processing-32k-length)
- [Contributors](#contributors)
- [Kanana 1.0](#kanana-10)
- [Citation](#citation)
- [Contact](#contact)
<br>
# Kanana 1.5
`Kanana 1.5`, a newly introduced version of the Kanana model family, presents substantial enhancements in **coding, mathematics, and function calling capabilities** over the previous version, enabling broader application to more complex real-world problems. This new version now can handle __up to 32K tokens length natively and up to 128K tokens using YaRN__, allowing the model to maintain coherence when handling extensive documents or engaging in extended conversations. Furthermore, Kanana 1.5 delivers more natural and accurate conversations through a __refined post-training process__.
#### ✨ Updated Kanana-1.5-15.7B-A3B
Introducing `Kanana-1.5-15.7B-A3B`, the first Mixture-of-Experts (MoE) model in our Kanana family, engineered for exceptional efficiency and powerful performance. `Kanana-1.5-15.7B-A3B`, which has sparse architecture, delivers capabilities comparable to the `Kanana-1.5-8B` dense model while utilizing only 37% of the FLOPS per token, making it a highly inference-efficient and cost-effective solution for real-world applications. Furthermore, `Kanana-1.5-15.7B-A3B` is powered by our newly enhanced post-training strategy, which includes on-policy distillation followed by reinforcement learning.
<!-- <p align="center">
<picture>
<img src="assets/performance/kanana-1.5-radar.png", width="700" style="margin: 40px auto;">
</picture> -->
> [!Note]
> Neither the pre-training nor the post-training data includes Kakao user data.
## Performance
### Base Model Evaluation
<table>
<tr>
<th>Models</th>
<th>MMLU</th>
<th>KMMLU</th>
<th>HAE-RAE</th>
<th>HumanEval</th>
<th>MBPP</th>
<th>GSM8K</th>
</tr>
<tr>
<td><strong>Kanana-Flag-1.5-32.5B</strong></td>
<td align="center">76.76</td>
<td align="center">61.90</td>
<td align="center">89.18</td>
<td align="center">73.17</td>
<td align="center">65.60</td>
<td align="center">81.50</td>
</tr>
<tr>Excerpt of 41,333 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:b6436d7ba442f0bb, llm:Repository description: 'Kanana: Compute-efficient Bilingual Language Models' — indicates development of bilingual language models focused on compute efficiency.
matched fp:b6436d7ba442f0bb, llm:Repository description: 'Kanana: Compute-efficient Bilingual Language Models' — indicates development of bilingual language models focused on compute efficiency.
matched fp:b6436d7ba442f0bb, llm:Repository description: 'Kanana: Compute-efficient Bilingual Language Models' — indicates development of bilingual language models focused on compute efficiency.