HTTP 200 verified twice
Finding your next stop…
Loading the latest directory information.
Loading the latest directory information.
Agents-A1 is a 35B Mixture-of-Experts agentic model from InternScience with 3B active parameters and a 262K context window. Open-sourced in Hugging Face Transformers format with native tool-use, function calling, and agentic reasoning across multiple domains.
Start here for the decision-making essentials: what Agents A1 does, who it is for, how it is accessed, and the first-party sources behind this profile.
See official pricing
Read-only public captures of Agents A1’s homepage and verified pricing page. Screenshots are dated, never live embeds, and open full-screen.
Published specifications and use cases for Agents A1. Repository dates are kept separate from the model's original release date.
Short answers to the questions buyers and builders commonly ask about Agents A1. Each answer cites the shared ledger below, where every source is listed once.
Compatible with Hugging Face Transformers, vLLM, and SGLang inference frameworks
These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, etc.
2026-06-26 (35B-A3B); 2026-07-14 (4B variant)
2026.6.26: 🔥🔥 We have open-sourced the Agents-A1 35B-A3B model
262144 tokens (262K)
--context-length 262144
35B total / A3B active
we have open-sourced the Agents-A1 35B-A3B model
Multimodal (text + vision encoder); supports a text-only mode that skips the vision encode
Text-Only (skips vision encoder to free KV cache memory)
Open-source (model weights and configuration files distributed in Hugging Face Transformer
This repository contains model weights and configuration files for Agents-A1 in the Hugging Face Transformers format.
This profile connects the jobs Agents A1 is described as handling with its delivery model, access options and the subjects used to match it to related products in this directory.
Each fact points to a recorded source, making it easy to distinguish verified product information from claims that need checking.
HTTP 200 verified twice
InternScience/Agents-A1 · Hugging Face
Model: Agents-A1 35B-A3B · Developer: InternScience · Architecture: Mixture-of-Experts (MoE)
Agents-A1 35B-A3B
InternScience
Mixture-of-Experts (MoE)
35B total / A3B active
Open-source (model weights and configuration files distributed in Hugging Face Transformers format)
2026-06-26 (35B-A3B); 2026-07-14 (4B variant)
262144 tokens (262K)
Multimodal (text + vision encoder); supports a text-only mode that skips the vision encoder
Natively supports function calling and tool integration with APIs, code interpreters, search engines, and external tools
A concise view of the jobs, capabilities and integrations described in the recorded product sources.
Documented product formats, platforms and official distribution destinations. Availability can vary by region and plan.
A concise history of software releases and material product changes. Events appear only when a dated source supports what changed.
Showing the newest updates and meaningful milestones. Open an entry for its summary and source.
The public model repository was created on Hugging Face. This repository date may differ from the model's original release date.
View source [1]Facts, answers, structured details, milestones and primary resource links cite this shared ledger. Each external page appears once; release tags from the same GitHub project are grouped under one release history.
These tools share a workflow, capability or audience with Agents A1. They may complement it rather than replace it, and are not presented as integrations or endorsements.
Agentic reasoning: decomposes complex tasks into executable sub-steps, plans ahead, and adapts strategy based on intermediate results
Quantized model variants released at multiple scales, including MLX versions from the mlx-community for Apple Silicon
Compatible with Hugging Face Transformers, vLLM, and SGLang inference frameworks
Model name: Agents-A1 35B-A3B · Architecture: Mixture-of-Experts (MoE) · Parameter count: 35B total, A3B active
Agents-A1
35B total parameters (A3B active)
2026-06-26 (35B-A3B open-sourced)
Hugging Face Transformers, vLLM, SGLang
Open-weights in Hugging Face Transformers format
Yes — open-sourced model weights, evaluation code, and technical report
Natively supports function calling and tool integration
Agentic reasoning — decomposes complex tasks, plans ahead, adapts strategy
Long-horizon Search, Engineering, Scientific Research, Instruction Following, Tool-calling
Model is Agents-A1, a 35B Mixture-of-Experts agentic model from InternScience · Architecture is Mixture-of-Experts (MoE) · Parameter count is 35B total with 3B active (35B-A3B), plus a 4B variant
Agents-A1
35B total / 3B active (35B-A3B); a 4B variant is also released
2026-06-26 for the 35B-A3B model; 2026-07-14 for the 4B model
Open weights and configuration files released in Hugging Face Transformers format; compatible with vLLM and SGLang
Natively supports function calling and tool integration with APIs, code interpreters, and search engines
Long-horizon Search, Engineering, Scientific Research, Instruction Following, and Tool-calling
Quantized model variants released via the Agents-A1 collection, with additional MLX-community versions for Mac
Can be served via SGLang or vLLM on 1 GPU at 262K context, exposed as an OpenAI-compatible endpoint at http://localhost:8000/v1
Model is Agents-A1 (35B-A3B) from InternScience · 35B Mixture-of-Experts architecture · 262,144 token context length in serving
Agents-A1 (35B-A3B) from InternScience, a 35B Mixture-of-Experts agentic model
35B parameters (35B-A3B MoE)
2026-06-26 (35B-A3B initial open-source release)
262,144 tokens (262K) supported in serving configurations
Native function calling / tool use (APIs, code interpreters, search engines, external tools)
