# Agents-A1: A 35B MoE Built for Agentic Scaling

> InclusionAI's new mixture-of-experts model bets that agent-horizon scaling can rival far larger systems on long-running tasks.

Published by The Open Weights on Jun 22, 2026. Canonical: https://theopenweights.com/news/agents-a1-ezpm

## Key facts

- Company: InternScience
- Model: Agents-A1
- Version: 1.0
- Category: Reasoning
- Modalities: Text / LLM, Reasoning
- License: Apache 2.0 (Open weights, commercial use allowed)
- Parameters: 35B (mixture of experts)
- Context window: 262K tokens
- Architecture: Qwen3_5MoeForConditionalGeneration
- Precision: BF16
- Weights size: 70.2 GB
- Significance: notable
- Published: 2026-06-22
- Last verified: 2026-07-22
- Hugging Face: https://huggingface.co/InternScience/Agents-A1
- Canonical URL: https://theopenweights.com/news/agents-a1-ezpm

InclusionAI has released **Agents-A1**, a 35-billion-parameter mixture-of-experts (MoE) model aimed squarely at agentic workloads — the kind of multi-step, tool-using reasoning tasks that increasingly define how AI systems get deployed in practice. The model is available now on [Hugging Face](https://huggingface.co/InternScience/Agents-A1) under a permissive Apache 2.0 license.

The headline claim is ambitious: that Agents-A1 reaches performance comparable to trillion-parameter systems through what its makers call *agent-horizon scaling*. Rather than simply growing parameter counts, the approach emphasizes a model's ability to plan and act over longer task horizons — a shift in where the gains come from.

## Why it matters

Most frontier-scale capability has come from raw size, which keeps the best models out of reach for teams without large compute budgets. A 35B MoE that punches above its weight on agentic benchmarks would be meaningful for a few reasons:

- **Efficiency**: MoE architectures activate only a fraction of parameters per token, keeping inference costs lower than the headline size suggests.
- **Open licensing**: Apache 2.0 allows commercial use and modification without the restrictions seen in some other open releases.
- **Focus**: Optimizing for agent horizons rather than general chat reflects where a lot of real-world demand is heading.

As with any launch-day performance claims, the trillion-parameter comparison will need independent verification on public benchmarks. For now, Agents-A1 is an interesting data point in the growing argument that smarter training and architecture — not just scale — can close the gap with the largest models.

## Get the model

- [Hugging Face](https://huggingface.co/InternScience/Agents-A1)

## Sources

- [InternScience/Agents-A1](https://huggingface.co/InternScience/Agents-A1) — Hugging Face, Jun 22, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "Agents-A1: A 35B MoE Built for Agentic Scaling", The Open Weights, Jun 22, 2026, https://theopenweights.com/news/agents-a1-ezpm