Agents-A1: Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

     📰  Tech Report    |    🏠  Home Page

---

🔥 News

• 2026.7.2: 🔥🔥 Based on Agents-A1, we have released a series of quantized model variants. Please refer to the Agents-A1 collection. Besides, we’d like to thank the mlx-community for providing quantized versions at multiple scales. Try running Agents-A1 on your Mac!

• 2026.6.26: 🔥🔥 We have open-sourced the Agents-A1 35B-A3B model, along with the evaluation code for selected domains and the technical report.

---

Agent-A1 is a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories with an average length of 45K tokens. Based on this, we train Agents-A1 with a three-stage recipe. First, we perform full-domain supervised fine-tuning to align the base model with broad agentic behaviors. Second, we train domain-level teacher models to capture specialized expertise in each domain. Third, we propose a multi-teacher domain-routed on-policy distillation with salient vocabulary alignment to improve knowledge transfer efficiency across different domains, unifying six heterogeneous domains into one deployable student model.

Highlights

• Agentic Reasoning: Agents-A1 excels at decomposing complex tasks into executable sub-steps, planning ahead, and adapting its strategy based on intermediate results.
• Tool Use: Natively supports function calling and tool integration, enabling seamless interaction with APIs, code interpreters, search engines, and other external tools.
• Long-Context Understanding: Handles extended conversations and documents with strong coherence and recall.
• Instruction Following: Precisely follows detailed, multi-constraint instructions across diverse domains.

We welcome developers and enterprises to integrate and try Agents-A1 and share their feedback.

Performance

We evaluate Agents-A1 in real-world agentic and research-oriented workflows across six directions — long-horizon search, engineering tasks, scientific research, instruction following, general agentic tasks, and scientific agentic tasks. Despite operating in the ~35B model class, Agents-A1 delivers highly competitive performance against frontier-scale systems such as GPT-5.5, DeepSeek-V4-pro, and Kimi-K2.6. It achieves overall SOTA results on several challenging benchmarks, including Seal-0 (56.4), HiPhO (46.4), FrontierScience-Olympiad (79.0), FrontierScience-Research (40.00), IFBench (80.6), and IFEval (94.8), while also ranking as the best among comparable models on a broad range of tasks such as BrowseComp (75.5), XBench-DS-2510 (86.0), GAIA (96.0), SciCode (44.3), HLE with tools (47.6), and MolBench-bind (56.8). These results show that Agents-A1 combines strong long-horizon search ability, robust scientific reasoning, and reliable instruction following, establishing it as a highly capable and efficient agentic model that narrows the gap with much larger frontier models.

🥇 Overall SOTA    🟢 Best Among Comparable Models (~35B)

Benchmark	📏 Comparable Models (~35B)			🚀 Larger-scale Models				⭐ Ours
	Qwen3.5-35B-A3B	Qwen3.6-35B-A3B	Nex-N2-mini	Step-3.5-Flash	Kimi-K2.6	DeepSeek-V4-pro(Max)	GPT-5.5(xhigh)	Agents-A1
🔍 Long-horizon Search
BrowseComp	61.0	67.93	74.1	69.0	83.2	83.4	🥇 84.4	🟢 75.51
XBench-DS-2510	77.0	71.0	82.0	56.3	🥇 90.0	🥇 90.0	84.0	🟢 86.0
Seal0	41.4	38.74	49.55	36.94	50.45	54.95	42.34	🥇 56.36
GAIA	59.8	78.64	82.52	84.5	80.58	🥇 98.06	87.38	🟢 96.04
⚙️ Engineering Tasks
SciCode	37.7	35.8	29.9	40.4	53.5	50.0	🥇 56.1	🟢 44.33
MLE-Lite	24.24	34.85	34.85	54.55	62.12	63.64	🥇 72.73	🟢 43.94
🧪 Scientific Research
HLE w/ tools	47.4	36.2	32.0	23.1	🥇 54.0	48.2	52.2	🟢 47.6
HiPhO	37.0	37.7	38.5	38.3	41.1	38.7	43.3	🥇 46.4
FrontierScience-Olympiad	64.5	60.3	52.0	61.0	73.0	76.0	78.0	🥇 79.0
FrontierScience-Research	2.5	2.9	5.0	6.7	17.9	13.3	26.7	🥇 40.0
📋 Instruction Following
IFBench	70.2	64.4	54.08	64.6	71.77	73.47	75.9	🥇 80.61
LongBench-v2	59.0	57.7	59.6	57.5	62.0	🥇 64.3	-	🟢 60.2
IFEval	91.9	91.3	88.4	93.53	94.45	93.35	93.35	🥇 94.82
🤖 General Agentic Tasks
τ2-Bench	🟢 81.2	79.0	74.53	75.77	81.93	🥇 82.2	81.63	79.81
VitaBench	31.9	35.6	23.0	30.0	35.63	🥇 49.04	45.0	🟢 38.75
🔬 Scientific Agentic Tasks
MatTools	21.0	15.9	34.1	44.93	63.8	47.1	🥇 68.8	🟢 47.1
MolBench-bind	46.0	48.7	51.4	45.95	21.6	37.8	🥇 62.2	🟢 56.8

Usage

SGLang

SGLang is a fast serving framework for large language models and vision language models.

Install SGLang with uv:

uv venv --python 3.12 --seed --managed-python
source .venv/bin/activate

uv pip install sglang

See its documentation for more details.

The following commands create API endpoints at http://localhost:8000/v1:

• Standard Version (1 GPUs, 262K context):

  python -m sglang.launch_server \
    --model-path InternScience/Agents-A1 \
    --port 8000 \
    --tp-size 1 \
    --mem-fraction-static 0.8 \
    --context-length 262144 \
    --reasoning-parser qwen3

• Tool Use:

  python -m sglang.launch_server \
    --model-path InternScience/Agents-A1 \
    --port 8000 \
    --tp-size 1 \
    --mem-fraction-static 0.8 \
    --context-length 262144 \
    --reasoning-parser qwen3 \
    --tool-call-parser qwen3_coder

vLLM

vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs.

Install vLLM from the main branch via uv:

uv venv --python 3.12 --seed --managed-python
source .venv/bin/activate

uv pip install vllm --torch-backend=auto

See its documentation for more details.

The following commands create API endpoints at http://localhost:8000/v1:

• Standard Version (1 GPUs, 262K context):

  vllm serve InternScience/Agents-A1 \
    --port 8000 \
    --tensor-parallel-size 1 \
    --max-model-len 262144 \
    --reasoning-parser qwen3

• Tool Call:

  vllm serve InternScience/Agents-A1 \
    --port 8000 \
    --tensor-parallel-size 1 \
    --max-model-len 262144 \
    --reasoning-parser qwen3 \
    --enable-auto-tool-choice \
    --tool-call-parser qwen3_coder

• Text-Only (skips vision encoder to free KV cache memory):

  vllm serve InternScience/Agents-A1 \
    --port 8000 \
    --tensor-parallel-size 1 \
    --max-model-len 262144 \
    --reasoning-parser qwen3 \
    --language-model-only

Note

For multi-GPU deployment, e.g., serving the model on 4 GPUs, please refer to the example launch script in the scripts/ directory.

Recommended Sampling Parameters

For the best generation quality, we recommend the following sampling parameters:

• temperature: 0.85
• top_p: 0.95
• top_k: 20
• min_p: 0.0
• presence_penalty: 1.1
• repetition_penalty: 1.0

Agent Capability Evaluation

To provide the community with a unified agent evaluation codebase for fair comparison, we have also open-sourced an evaluation framework for assessing agentic models across core capabilities, including tool use and multi-step reasoning. The evaluation code is included in the evaluation/ of this repository.

We use this framework to evaluate the released model under a standardized and reproducible setting. Specifically, the model is tested on a set of agent-oriented tasks that require it to understand user goals, decompose complex instructions, interact with tools or environments when necessary, and produce final results. The evaluation results reported in Model Card are generated using the open-source framework above, so that users can reproduce the experiments, compare other models under the same protocol, and further extend the benchmark for new agent scenarios. (Note that: To ensure a fair comparison, we report the benchmark results from their original technical reports. If a model does not report the corresponding benchmark results, we evaluate it using the same evaluation protocol as our model.)

For detailed evaluation scripts, task definitions, metrics, and reproduction instructions, please refer to the evaluation codebase.

Citation

If you find our work helpful, feel free to give us a cite.