@harvey
Account based inUnited States
About this account
- Account based in
- United States
- Connected via
- Web
Account-level information from X, not a live location or the device used for a specific post.
AI for the world’s most complex legal work.
Joined March 2023
- Tweets590
- Following3
- Followers22.1K
- Likes326
Pinned Tweet
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.
Review Tables now load up to 5.8x faster in Harvey.
Our engineering team built server-side pagination to support larger reviews without loading the entire table into browser memory.
That required:
- A Turbopuffer query layer for server-side sorting and filtering
- Custom row-ID pagination and staged loading
- Typed and string fields to handle inconsistent model outputs
- Fallback reads and backfill jobs to recover from data drift
Read our engineering blog on scaling Review Table UI: harvey.ai/blog/making-review…
Harvey retweeted
Curious how @harvey makes hundreds of tailored demo videos?
Through Figma’s MCP, agents work directly on the canvas to:
→ Pull in visual context from their brand and motion guidelines designed in Figma
→ Storyboard and animate using Figma Motion, for review and feedback with their team
→ Generate imagery and video with @figmaweave
How @ShawnFarsai uses @figma @ElevenLabs and @claudeai to create hyper personalized marketing videos with agents
Harvey retweeted
Awesome to see @ShawnFarsai and @gabepereyra share Harvey's work around combining agentic workflows with the Figma canvas and Weave!
Excited to make this something you can do end to end in Figma! (We're so close!)
How @ShawnFarsai uses @figma @ElevenLabs and @claudeai to create hyper personalized marketing videos with agents
Harvey retweeted
how to channel your ai psychosis and creativity into arr featuring @Figma and @ElevenLabs
How @ShawnFarsai uses @figma @ElevenLabs and @claudeai to create hyper personalized marketing videos with agents
How Harvey's marketing org is using @figma, @ElevenLabs and @claudeai to:
- Create entire launch films with agents
- Build tailored demo videos for each customer using sales calls and account insights
- Script and voice hundreds of films overnight
While burning 15 billion tokens over 2 weeks.
Deep dive by @gabepereyra and @ShawnFarsai:
How @ShawnFarsai uses @figma @ElevenLabs and @claudeai to create hyper personalized marketing videos with agents
Harvey retweeted
@JensenHuang at GTC:
"Here's Trajectory and Harvey using Nemotron, post-trained it and created their own frontier-level capability."
"With customization and with partners in the ecosystem that we work with, you could turn these models into the frontier.
And a result, turn your proprietary data into sovereign intelligence"
It's great to be doing work with such amazing companies like @harvey and @nvidia over here at @trajectorylabs!
A closer look at Harvey Graph.
Starting with graphs for US Case Law and Spaces, Harvey Graph maps the connections between matters, documents, and legal knowledge so agents can start work with the context they need.
In early Legal Agent Bench tests, adding a context graph raised rubric pass rates on firm-knowledge questions from 65% to 80%.
We built context graphs to help agents work through long-horizon legal tasks.
Starting with Legal Agent Bench tasks in 11 categories like M&A diligence and contract negotiation, we had agents build context graphs of the underlying documents (without access to the eval tasks or grading rubrics).
Context graphs dramatically improved agent performance on LAB tasks:
- Firm Knowledge tasks: 65% → 80%
- Precedent lists tasks: 67% → 80%
- Matter counting tasks: 53% → 83%
They also boost agent efficiency: we found a 48% cost reduction for agents given access to a context graph to assist their work.
Full video below - more to come soon.
An overview of how we think about context graphs at @harvey.
Context graphs are a promising direction for improving agent quality and efficiency. We ran a series of experiments on Legal Agent Bench to evaluate how structured context improves online agent performance.
For each LAB task we evaluated, an agent first reviews each document in its client matter and builds a graph of relevant entities and their relationships – e.g. parties, dates, draft vs. executed status, which versions belong to the same contract.
To evaluate the impact of the graph, we compared the performance of an agent with and without the context graph, using LAB’s standard grading setup. Overall, the agent with the context graph improved rubric pass rate by up to 30% while reducing task cost by up to ~50%.
In some cases the graph led to massive improvements. For example, in one firm-knowledge task, the agent was asked to find the firm’s most recent secured financing and retrieve the executed credit and security agreements as precedents for a new facility.
An agent working from raw files attempted to search through several files sequentially, and looked at many different matters, but ultimately selected the wrong precedent (0% of rubric criteria). With the context graph, the same agent found the relevant matter and its executed agreements immediately (75% of rubric criteria).
We’ll have more to share soon on how we’re building this into Harvey, along with more experiments like this one!
