@paulgug@0xneoptr@Franzferdinan57@Chris_Wozniczek Actually, I’m more curious about what triggered such a long reasoning process. I didn’t do anything special, just plugged the model into the DeepSeek harness and got fast, concise answers. I didn’t even change the reasoning parameters.
@flaviusburca@Tech2Wild Could you share more details about the evaluation, such as inference parameters and the mean and variance across multiple runs?
Sage A1.5 🔬 New agentic model built for long horizon scientific problem sloving
- 35B MoE
- 256k context
- Parallel subagents
- Built in self-verification
The model is open source. We’d love to hear what you build and where it can improve!
Hugging Face:
https://t.co/5DEiyfFHyy
Try Sage-A1.5:
https://t.co/Wm50bwIn1N
Since Agents-A1 launched, it and its community derivatives have surpassed 1 million downloads.
Now, we’re introducing Sage-A1.5, the next step in bringing stronger coding and scientific research capabilities into one agentic model.
Sage-A1.5 is here, bringing 256K context, parallel research, and evidence checks to long-horizon scientific workflows.
🤖 https://t.co/yiBBWUlG7K
🏆 SOTA across scientific benchmarks: Scores 90.0 on PinchBench, 81.0 on FrontierScience-Olympiad, and 48.3 on FrontierScience-Research, ahead of Kimi-K3, GLM-5.3, and GPT-5.6-sol in the reported comparison.
🧩 Agentic Orchestration: When enabled, delegates independent investigations to parallel subagents with isolated contexts, then integrates their evidence and findings to guide subsequent work.
✅ Verification: Optional Verification Mode reviews progress, checks evidence and assumptions, and redirects stalled searches within the model's own reasoning, without a separate critic model at inference.
🔬 Open Scientific Discovery: Supports exploratory research through tool-assisted analysis and produces inspectable research artifacts, extending its use beyond predefined question answering.
🛠️ Native tool use for research: Supports APIs, code interpreters, search engines, and external tools for tool-augmented scientific reasoning and professional workflows.
⚡ Available in BF16, FP8, and GGUF F16/Q8/Q4 formats, with Transformers, vLLM, and SGLang support.
📜 Apache 2.0.
In electrochemistry, Sage-A1.5 integrated 16 papers with experimental feedback to refine reaction conditions.
It recovered the known optimal conditions yielding 89% isolated yield and proposed mechanistic hypotheses for further experimental testing.