AXCXEPT Open Models: Model List and Licenses
About AXCXEPT
AXCXEPT (AXCXEPT co., ltd.) is a Japanese company. On Hugging Face it publishes models from an individual account (AXCXEPT) rather than an organization.
It releases strong overseas models further trained for Japanese, in the EZO and Borea series, built on a wide range of bases including Qwen, Gemma, Llama, Phi and InternVL.
What It Releases
- Qwen3-EZO-8B-beta (May 2025): an 8B model based on Qwen3-8B. According to its model card, it scored 9.08 on MT-Bench and 8.87 on the Japanese JMT-Bench, comparable to Gemini 2.5 Flash and GPT-4o on multi-turn tasks. The model card itself notes that this was an internal evaluation with GPT-4o and Gemini 2.5 Flash as judges, and results may differ under external or official conditions.
- The EZO series (2024 to 2025): models based on Qwen2.5 (32B, 72B), Gemma 2 and 3, Llama 3.2, InternVL2 and others.
- Borea (August 2024): task-specific models (Japanese, coding, general) based on Phi-3.5-mini.
- No new models have been released since May 2025 (as of 2026-09-26).
Licenses
Models inherit the license of their base model. Qwen-based models are Apache-2.0, Phi-based models are MIT, and Gemma- and Llama-based models use their original licenses. Read each model’s license text before use. This summary is not legal advice.
Running Them Locally
- There are no official GGUF builds (as of 2026-09-26); look for third-party conversions.
- On a consumer GPU, Qwen3-EZO-8B or the 3.8B Borea models are easy sizes.
Sources: the model card of AXCXEPT/Qwen3-EZO-8B-beta; the model list of the AXCXEPT account on Hugging Face (all as of 2026-09-26).
Our Coverage and Data
Local Model Watch has not published an article on AXCXEPT’s models yet. The model list below is collected by code from Hugging Face every day. Part of our publisher index; model families link to their family pages.
At a Glance
| Item | Value |
|---|---|
| Hugging Face organization | AXCXEPT |
| Models on Hugging Face | 25 |
| Articles on its own models | 0 |
| Articles on derived models | 0 |
Recent Models on Hugging Face
The publisher lists 25 models in total on Hugging Face. Below are up to 15 of the most recent generative models (text, image, video, audio and so on), the scope of this site, per organization. Classifiers, feature extractors, research components and repositories without a model card are omitted, and the publisher’s own quantized builds and helper variants are folded into the row of the original model. Parameter counts are left blank for repositories that pack weights into integer types, where Hugging Face’s count is not the real parameter count.
→ Scroll horizontally to see all columns
| Released | Model | Task | Parameters | License | Official variants |
|---|---|---|---|---|---|
| 2025-05-11 | AXCXEPT/Qwen3-EZO-8B-beta | text generation | 8.2B | apache-2.0 |
— |
| 2025-04-14 | AXCXEPT/EZO2.5-gemma-3-12b-it-Preview | vision-language | 12.2B | gemma |
— |
| 2025-04-09 | AXCXEPT/QwQ-32B-Distill-Qwen-1.5B-Alpha | text generation | 1.8B | mit |
— |
| 2025-01-28 | AXCXEPT/phi-4-deepseek-R1K-RL-EZO | text generation | 14.7B | mit |
— |
| 2025-01-26 | AXCXEPT/phi-4-open-R1-Distill-EZOv1 | text generation | 14.7B | mit |
— |
| 2024-10-03 | AXCXEPT/EZO-gemma-2-2b-jpn-it | text generation | 2.6B | gemma |
— |
| 2024-09-30 | AXCXEPT/llm-jp-3-3.7b-instruct-EZO-Humanities | text generation | 3.8B | apache-2.0 |
— |
| 2024-09-30 | AXCXEPT/llm-jp-3-3.7b-instruct-EZO-Common | text generation | 3.8B | apache-2.0 |
— |
| 2024-09-26 | AXCXEPT/EZO-Llama-3.2-3B-Instruct-dpoE | text generation | 3.2B | llama3.2 |
— |
| 2024-09-24 | AXCXEPT/EZO-AutoCoTRAG-Qwen2.5-72B-Instruct_q4 | text generation | — | apache-2.0 |
— |
| 2024-09-24 | AXCXEPT/EZO-AutoCoTRAG-Qwen2.5-32B-Instruct | text generation | 32.8B | apache-2.0 |
— |
| 2024-09-21 | AXCXEPT/EZO-Qwen2.5-32B-Instruct | text generation | 32.8B | apache-2.0 |
— |
| 2024-09-20 | AXCXEPT/EZO-Qwen2.5-72B-Instruct | text generation | 72.7B | apache-2.0 |
— |
| 2024-08-21 | AXCXEPT/Borea-Phi-3.5-mini-Instruct-Coding | text generation | 3.8B | mit |
— |
| 2024-08-21 | AXCXEPT/Borea-Phi-3.5-mini-Instruct-Common | text generation | 3.8B | mit |
— |
From the Hugging Face API. Last changed 2026-09-27 (JST). Collected by code.
Last updated 2026-09-27 (JST). The explanation at the top of this page was written with the help of AI from the primary sources it cites. The tables and lists under “Our Coverage and Data" are assembled by code from our article log and the Hugging Face API.