-
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
Paper • 2610.28578 • Published • 55 -
InternScience/Sage-A1.5
Text Generation • 36B • Updated • 6 • 9 -
InternScience/Sage-A1.5-FP8
Text Generation • 36B • Updated • 3 -
InternScience/Sage-A1.5-Q4_K_M-GGUF
Text Generation • 36B • Updated • 5
AI & ML interests
None defined yet.
Recent Activity
Papers
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering
Organization Card
Welcome to the InternScience organization, the open-source hub of AI for Science Center @Shanghai AI Laboratory. We focus on open-sourcing high quality LLMs/MLLMs, tools, and datasets to accelerate scientific discovery across disciplines.
Agents-A1 is a Long-horizon Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon.
-
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
Paper • 2606.30616 • Published • 104 -
InternScience/Agents-A1
Text Generation • 35B • Updated • 12.8k • 657 -
InternScience/Agents-A1-FP8
Text Generation • 35B • Updated • 499 • 24 -
InternScience/Agents-A1-Q4_K_M-GGUF
Text Generation • 35B • Updated • 21.2k • 151
-
Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition
Paper • 2610.28578 • Published • 55 -
InternScience/Sage-A1.5
Text Generation • 36B • Updated • 6 • 9 -
InternScience/Sage-A1.5-FP8
Text Generation • 36B • Updated • 3 -
InternScience/Sage-A1.5-Q4_K_M-GGUF
Text Generation • 36B • Updated • 5
Agents-A1 is a Long-horizon Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon.
-
Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent
Paper • 2606.30616 • Published • 104 -
InternScience/Agents-A1
Text Generation • 35B • Updated • 12.8k • 657 -
InternScience/Agents-A1-FP8
Text Generation • 35B • Updated • 499 • 24 -
InternScience/Agents-A1-Q4_K_M-GGUF
Text Generation • 35B • Updated • 21.2k • 151
spaces 5
pinned
Running
Agents
9
SGI-Bench Leaderboard
🥇
Scientific General Intelligence of LLMs/vLLMs
pinned
Paused
Agents
4
SciEval Leaderboard
🥇
Open, science-focus leaderboards benchmarking LLMs and VLMs
Sleeping
7
ResearchHarness
🚀
Lightweight harness for tool-using LLM agents.
Running
Agents
1
ResearchClawBench Task Submission
📦
Validate and submit a ResearchClawBench task ZIP to Hugging Face
models 21
InternScience/Sage-A1.5-F16-GGUF
Text Generation • 36B • Updated • 1
InternScience/Sage-A1.5-FP8
Text Generation • 36B • Updated • 3
InternScience/Sage-A1.5-Q8_0-GGUF
Text Generation • 36B • Updated • 2
InternScience/Sage-A1.5-Q4_K_M-GGUF
Text Generation • 36B • Updated • 5
InternScience/Sage-A1.5
Text Generation • 36B • Updated • 6 • 9
InternScience/MLExpert
Text Generation • 35B • Updated • 543 • 14
InternScience/Agents-A1
Text Generation • 35B • Updated • 12.8k • 657
InternScience/Agents-A1-F16-GGUF
Text Generation • 35B • Updated • 1.37k • 14
InternScience/Agents-A1-Q8_0-GGUF
Text Generation • 35B • Updated • 638 • 37
InternScience/Agents-A1-Q4_K_M-GGUF
Text Generation • 35B • Updated • 21.2k • 151
datasets 23
InternScience/ResearchClawBench
Benchmark • Updated • 57 • 34.3k • 16
InternScience/Agents-A1-ifstruct-eval-20260703-041434
Preview • Updated • 34 • 1
InternScience/Scholar-kg
Updated • 35 • 9
InternScience/SGI-Reasoning-Lite
Viewer • Updated • 72 • 17 • 1
InternScience/SGI-DryExperiment-Lite
Viewer • Updated • 67 • 16 • 1
InternScience/SGI-IdeaGeneration-Lite
Viewer • Updated • 78 • 15 • 1
InternScience/SGI-DeepResearch-Lite
Viewer • Updated • 60 • 16 • 2
InternScience/SGI-DeepResearch-Gold
Viewer • Updated • 243 • 33 • 4
InternScience/SGI-Reasoning
Viewer • Updated • 291 • 82 • 8
InternScience/SGI-WetExperiment
Viewer • Updated • 68 • 58 • 3