Ryan Prudhomme retweeted
Harvey Q3 Recap:
Metrics:
- Added over $100M ARR and crossed $400M ARR
- 97% GRR
- 20% of the Fortune 500 and 40% of the Fortune 50 use Harvey
- 80% of the Am Law 100 use Harvey
Key hires:
- @asadovsky, Chief Research Officer (former CVP AI, Microsoft AI & Distinguished Engineer, Google DeepMind)
- @stevezad, Chief Revenue Officer (former GVP Rubrik)
Product:
- Harvey II
- Horizon Scanning
- Command Center GA
Engineering:
- Agent harness end-to-end redesign
- Brought agents to Word, Outlook, Vault, and Review Tables
- Vault files 2x+ in Q3, token throughput 20x+ since January, vector DB migration, major cost & efficiency gains
AI:
- Harvey Tenet, our first model post-trained for legal
- Legal Agent Bench extensions for M&A Diligence and Firm Knowledge
- Research collaborations with @FireworksAI_HQ, @appliedcompute, @baseten, @EngramLab, and @trajectorylabs
Ryan Prudhomme retweeted
We’ve raised $550M at a $15.5B valuation co-led by @lightspeedvp and @DiffusionVC.
We're using this funding to help law firms, in-house legal teams, and professional services build and own their intelligence.
Ryan Prudhomme retweeted
Tenet is @harvey's first model post-trained for legal
It reaches frontier performance on Legal Agent Bench and generalizes to other agentic benchmarks
Tenet comes with three subagents post-trained for specialized tasks: M&A diligence, review tables, and firm knowledge
It's Harvey's first major step towards building frontier legal intelligence and enabling firms to build their own specialized models
More from @gabepereyra:
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.
Ryan Prudhomme retweeted
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.
So much to love about this video but this is my favorite part.
Introducing Harvey II: smarter agents from the start.
- Featuring Harvey Tenet, our first model trained for legal work
- Built around matters and projects
- Agents start with the files, context, permissions, and history they need
- Assign tasks to lawyers or agents, then track and review the work
- Harvey remembers how you work and writes like you
Ryan Prudhomme retweeted
been a bit of a week and appreciate yall as I go through my DMs!
as stated we're looking to bring on a Product Designer and Frontend Eng to our Design Team at Parallel.
iso infinitely curious Internet and AI natives 👀🌐
Ryan Prudhomme retweeted
We’re training a series of legal foundation models at Harvey.
Excited to launch Harvey Research and invest heavily in model training, building on our Legal Agent Benchmark and research collaborations.
Introducing Harvey Research:
We've shared our model strategy.
We've open-sourced Legal Agent Benchmark, the largest benchmark for long-horizon legal work spanning 1,200 tasks across 24+ practice areas.
And we've collaborated on research with leading neolabs and inference providers like @baseten, @trajectorylabs, @LangChain, @FireworksAI_HQ, @appliedcompute, and @EngramLab.
Now we have a home base for it.
Live at: harvey.ai/research
Ryan Prudhomme retweeted
Excited to welcome Alec, Connor and the Benchmark team to Harvey.
harvey.ai/blog/y-combinator-…
Ryan Prudhomme retweeted
Come help us scale @harvey’s model training team.
If you’re interested in bringing frontier agent research into the Harvey product and working with:
- @baseten to scale up RL to 80M+ token virtual datarooms
- @PrimeIntellect to create structured agent training environments from unstructured legal data
- @FireworksAI_HQ to navigate the quality <> cost Pareto frontier with inference-time routing and advisor models
- @LangChain & LangChain Labs to build efficient verifiers and close the observability <> training feedback loop
- @appliedcompute to post-train open weight models and high-volume agents for end-to-end legal tasks
- @EngramLab to create an entire synthetic law firm and firm knowledge memory systems for better / more efficient open-world search
- @trajectorylabs & @NVIDIAAI to shape the frontier of continual learning and sovereign AI for high-stakes domains
- @mercor & @SnorkelAI to build out Legal Agent Bench and other benchmarks across legal and other verticals
and other projects like this, then this is the role for you.
Apply here: harvey.ai/company/careers/d7…
We are hiring for @Harvey’s model training team.
This team will help Harvey expand from the application layer into the model layer and from legal into high end knowledge work more broadly.
We are hiring AI researchers of all seniority, particularly those with experience post-training frontier or open source models.
Our program is centered around large-scale model training, synthetic data generation, long horizon reinforcement learning, and rigorous evaluation in real world deployments. We are scaling-pilled and believe that nothing beats the combination of larger models and better training data.
We’ve been able to generate incredibly realistic legal environments and validated that this allows us to post-train open source models to achieve frontier performance with agents. We plan to scale up these data generation and training efforts significantly across legal to start, and eventually other verticals.
As a researcher, you will have access to thousands of GPUs and unique training data from our product and customer relationships.
Your research will inform Harvey’s product strategy and power AI used for some of the most economically and societally impactful work in the world.
Ryan Prudhomme retweeted
Mayor Mamdani Delivers Address Marking America’s 250th Birthday nitter.gq/i/broadcasts/1aJbddvRv…
Ryan Prudhomme retweeted
Would Harvey Specter use Harvey?
@GabrielMacht had to ask.