0

Vals AI

Enterprise-focused LLM evaluation company benchmarking models on real-world legal, finance, and tax tasks.

Visit
Type
eval co
HQ
New York, NY, USA
Founded
2023
Funding
Seed-stage; ~$2M raised 2024 (Lightspeed, Y Combinator W24)
Website
vals.ai
GitHub
github.com/vals-ai
Twitter
twitter.com/vals_ai

Cite

Notes

Only stored in your browser.

Evals
12
Tools
0
Models
0
Papers
0
Boards
1
People
2

Leaderboards

1
Vals AI LegalBenchVals AI's continuously updated leaderboard for legal-domain LLM tasks, grounded in the academic LegalBench benchmark plus Vals's own attorney-graded private evaluations.

Evals

12

Vals.ai Legal Evals

Vals.ai's proprietary suite of legal-domain benchmarks (contract review, hallucination tests, LegalBench Pro) used by law firms to procure LLMs.

ActiveLegal ReasoningHallucinationFactual RecallLegal

TaxEval v2

A Vals-created set of questions and responses to tax questions

Finance

Vibe Code Bench v1.1

Can models build web applications from scratch?

Coding

Finance Agent (v2)

Evaluating agents on core financial analyst tasks using the FAB v2 harness

Finance

MortgageTax

Evaluating reading and understanding tax certificates as images

Finance

MedCode

Can models support the medical billing process?

Medical

MedScribe

Can models support doctors with their administrative work?

Medical

Vals Index

Benchmark consisting of a weighted performance across finance and coding tasks. Showing the potential impact that LLM's can have on the economy.

Vals Multimodal Index

Benchmark consisting of a weighted performance across finance, coding, and education tasks. Showing the potential impact that LLM's can have on the economy.

Multimodal

CorpFin v2

A private benchmark evaluating understanding of long-context credit agreements

Finance

SAGE

Student Assessment with Generative Evaluation

Legal

ProofBench

Automated theorem proving benchmark

Mathematics