One of the underrated pieces of our post-training work with @baseten is exploring KV cache compaction.
In our training runs over Legal Agent Bench, agents converged to similar policies, roughly:
1. Ingest client matter documents up front
2. Analyze + iterate on
Now live in Harvey: Claude Sonnet 5.
Sonnet 5 scored 5.8% all-pass on our Legal Agent Benchmark (LAB), building on Sonnet 4.6 with broad gains in legal accuracy and output quality.
Inference-time model routing based on legal practice area dramatically improves agent performance.
So do other types of "blended intelligence":
- collaborative model teams,
- advisor-executor patterns,
- and model routing based on inferred user preference.
While these methods
Harvey is now the exclusive Legal AI partner of the Toronto Maple Leafs.
We're proud to partner with one of the most iconic franchises in sports as we continue investing in Toronto and expanding our presence across Canada.
Harvey FORUM comes to Sydney, Australia, September 2, 2026.
One day in Sydney, bringing together leaders from across APAC for expert perspectives, peer dialogue, and the conversations reshaping legal.
Save the date: more details coming soon.
We partnered with @appliedcompute to train a legal agent.
We optimized each part of the agent stack:
- the evaluation loop
- the agent harness
- and post-trained the underlying GLM-5.1 model.
The result? The agent achieved the highest rubric pass rate on our Legal Agent
NEW: Harvey Co-Founder + Head of Applied Research on the *Token Reckoning*
Valued at $11B, @harvey is on a mission to win the entire legal category, competing head-on against the trillion-dollar labs
Coding agents hit Karpathy's "agents work now" inflection in late 2025. Harvey
Visualize legal data on demand in Harvey.
Ask Harvey to build interactive timelines, entity charts, and compliance dashboards from your inputs - filterable, shareable, and ready in minutes.
We're training the first in a series of legal foundation models.
Harvey Labs is hiring across the post-training, agent, and data stack. We're also interested in acquiring teams and neolabs with talent in these areas.
DM @gabepereyra if you're excited by frontier legal AI.
Model strategy for @harvey:
We are working on the first model in our legal foundation model series, inspired by @cursor_ai's Composer. Two goals:
1. Allow us to serve frontier intelligence across our product surface areas at an affordable price and a strong security posture.
