● Covered by 1 source · 1 reportMedium impact1 neutral
Inkling 2 Small Mixture-of-Experts Model Released with Enhanced Agentic Capabilities
Key points
- Inkling 2 Small is a 594B parameter Mixture-of-Experts model.
- It has 17B active parameters and supports image inputs.
- Scores 31 on the Artificial Analysis Intelligence Index, outperforming prior models.
- Initially available via Tinker; open-source release planned for next week.
Inkling 2 Small Model Introduction
Inkling has introduced its second-generation model, Inkling 2 Small. This is a Mixture-of-Experts model featuring 594 billion total parameters and 17 billion active parameters. The model supports image inputs and offers variable effort capabilities.
Performance and Capabilities
Inkling 2 Small achieved a score of 31 on the Artificial Analysis Intelligence Index, indicating performance gains over previous models in areas such as agentic coding, vision, knowledge work, and reasoning. These improvements stem from advancements in the training process, including more efficient pre-training and a large, unified Reinforcement Learning (RL) run.
The model's vision and agentic capabilities allow it to use visual feedback for iterative development of websites, games, and 3D designs with interactive elements. It also demonstrates enhanced long-horizon autonomous work.
Agentic Coding and Autoresearch
Inkling 2 Small was trained using RL on hundreds of thousands of long-horizon coding tasks. The training incorporated an agentic RL grader that evaluated the model's problem-solving approach, not just code pass rates. Early checkpoints were tested on OpenCode under the name Fledge-Alpha.
The model is designed to be more effective at coding and easier to integrate into agent loops, following instructions better, considering changes, and verifying its work autonomously. Beyond conventional software, it was also trained on long-horizon autoresearch tasks, including agentic GPU kernel optimization and autonomous machine learning model post-training across various GPU architectures.
Availability and Licensing
Inkling 2 Small is currently accessible through Tinker with closed weights. The company plans to release the full weights under an Apache 2.0 license next week, following external safety testing and red-teaming. Users can try Inkling 2 Small on the Tinker Playground and it is available for fine-tuning and sampling on Tinker; checkpoints will be downloadable upon the release of the weights.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Reporting from
Inkling has released Inkling 2 Small, a 594B parameter Mixture-of-Experts model with 17B active parameters, supporting image inputs and variable effort. This new model shows improved performance in agentic coding, vision, knowledge work, and reasoning, surpassing previous Inkling models on the Artificial Analysis Intelligence Index with a score of 31. Inkling 2 Small is initially available through Tinker with closed weights, with plans for an Apache 2.0 open-source release after safety testing.