@AfterQuery just moved an open model past the cost-accuracy frontier in finance.
We post-trained @NVIDIAAI Nemotron 3.5 Super VL on expert credit tasks with @hebbia.
Perfect solved: 47% to 74%. That beats Sol 6, and lands within 3 points of Astra 6, at about 1/7 the cost.
Link to the full case study in replies.
Not all coding data is created equal.
Using only 500 tasks from @AfterQuery's SWE agent dataset, our internal post-training team was able to improve Qwen3.8-27B-Medium by 11.3 points in only 15 steps of GRPO.
Read more about our training experiment in the blog post linked in the comments.
Working on AI research, post-training, or deployment? Come meet the @AfterQuery Team during SF Tech Week.
Wed 10/7: Frontier of Finance with Goldman Sachs
Fri 10/9: Rooftop happy hour
We'd love to see you there. Details + RSVPs below.
People still underestimate how large the data market will become.
Better models do not reduce demand for data. They increase the number of useful things models can learn.
The frontier keeps moving, so the data has to keep getting harder.
Muse Spark 1.3 is rolling out today with frontier performance almost too cheap to meter. This is the biggest jump we've made so far on coding and agentic work. Try it in Muse Code and our API.
Next up 🍉 and Muse Spark open weights releases coming soon.
Legal and finance are two of the highest-impact domains to crack.
3.8 Flash now leads both Harvey’s Legal Agent Benchmark and Vals Finance Agent v2. Huge congrats to GDM ⚡️
Introducing Gemini 3.8 Flash, another jump in Gemini's agentic + coding capabilities, and our 3rd updated Flash model in only 6 weeks...
This model has been a ton of fun to work with, excited to see what you all think!
This is a historic moment for Korean sovereign AI and the open weights ecosystem writ large. We look forward to supporting Motif and the Korean AI ecosystem as they push the frontier of open intelligence.
Motif 3 leads its comparison set on τ³-Banking, and posts 94.7 on τ²-Bench Telecom, 74.9 on Terminal-Bench 2.1, and 76.2 on SWE-bench Verified. It’s also great to see @nvidia's open foundation put to use here - Motif 3's post-training was done via NeMo-RL.
In a world where it’s easy to assume that the AI race is already won, the team at Motif is proof that the field is open, and we’re still at the starting line.
591 Followers 2K Followingdata for robotics and health unlock the future we need. founder, https://t.co/ji2txiGF8Q, https://t.co/7eH2e6LVXg, https://t.co/AciEw5Exhq
30 Followers 719 FollowingBuilding AI agents that know when not to act. Classical agent design + LLMs, evals, human-in-the-loop. Open source: agent_design_peas. Finance & ops automation.
1K Followers 2K FollowingProduct @Microsoft AI | Reasoning agents, evals, RL.
UIUC alum.
"you can just do things"
Nerding out on Robots, AI & Longevity
Views & opinions my own.