AI for the world’s most complex legal work.
- Tweets 616
- Following 3
- Followers 22,559
- Likes 332
Introducing Tenet, our first model post-trained for legal.
Tenet is a Kimi K3 base that we post-trained with @FireworksAI_HQ on a corpus of publicly available legal data, synthetic data, and human expert data simulating long-horizon legal work.
Training increases Tenet's all-pass rate by 82% on LAB and 22% on LAB Contracts relative to the Kimi K3 base model. It achieves state-of-the-art performance on LAB Contracts and places second on LAB.
These gains generalize to other leading agentic benchmarks including @mercor's Apex Agents - Corporate Law, @crosbylegal's Redline Bench, and @scale_AI's Professional Reasoning Bench.
Tenet is also optimized for token efficiency, operating at less than a fourth the cost of leading foundation models.
We additionally post-trained three specialist models for Tenet to use as subagents:
1) M&A Diligence: post-trained with @baseten on our LAB Diligence environment in an RLM harness, this model is optimized for high-scale, long-horizon tasks.
2) Review Tables: trained with @appliedcompute on our Review Table environment, this model is state-of-the-art and cost-effective at high-volume document review and structured data extraction.
3) Firm Knowledge: trained with @EngramLab on our synthetic law firm environment, this model is optimized to learn and search over a firm's knowledge via memory and structured notes.
More details on model training, environment design, benchmarking, results, and more in the article by @gabepereyra below.
What's next for Harvey’s research?
- Scaling LAB to more jurisdictions, practice areas and workflows
- Scaling compute to bring new generalist models and capabilities to Harvey
More to come soon.
121
287
3,091
3,575,099
@katieburkie and John Haddock explore the adoption patterns across seniority levels, firm sizes, and markets.
Learn more: harvey.ai/blog/legal-ai-adop…
1
5
445
When a judge raised a pivotal question during a major tax dispute, a @ClaytonUtz lawyer found the answer in Harvey in just 40 seconds.
See how the firm uses Harvey to make critical information accessible when it matters most: link.harvey.ai/uugci
2
12
1,857
Harvey now enforces ethical walls using client matters and policies synced from @imanageinc Security Policy Manager (“SPM”).
Harvey respects the information barriers a firm already has, so lawyers can collaborate without a second set of rules to manage.
Read more: harvey.ai/blog/ethical-walls…
4
2
19
1,709
What we found:
A small group of frontier models can now research and write complete, first-draft briefs in a single shot, moving substantially closer to the high bar for filing with a court. But even the strongest still fall short of that bar, with issues like hallucinations and reliance on decisions that have been set aside.
1
1
376
Next, we plan to open-source our environment-building pipeline and basic pairwise grader, enabling researchers to construct their own case law environments and study these capabilities.
Full findings and methodology: link.harvey.ai/uufZD
2
344
Today @AnthropicAI added a gallery of ready-to-use artifacts to Claude for Legal, and Harvey is proud to be a launch partner.
Harvey Intake now connects Claude to your legal team through Harvey’s MCP server.
Business users can now submit requests in Claude while Legal assigns, tracks, and manages the work in Harvey.
Read more: link.harvey.ai/uubZP
9
24
243
48,556
Harvey is now FedRAMP Class D (High) certified.
U.S. federal agencies, contractors, and partners can now use Harvey for legal work in an environment built for sensitive government information.
Read the announcement: link.harvey.ai/uucoF
3
9
50
9,886
Horizon Scanning is now available in Harvey.
Monitor 11,000+ legislative and regulatory primary sources across 100+ jurisdictions in one feed.
See the developments most relevant to your business first, tagged by topic, jurisdiction, and date.
Draft emails, memos, and briefings from any finding, and use scans in Assistant, agents, and Spaces.
5
5
44
6,139