Hebbia
An AI document analysis platform that uses large language models to help professional services fields such as finance and law with document retrieval and analysis.
Timeline (last 90 days)
Hebbia says Astra can produce highly precise outputs and deliverables and follows the specifications set by users.
After testing OpenAI's Astra model, the Hebbia team saw a double-digit performance improvement on the citation source selection task compared with the other models they tested.
Hebbia defines "citation source selection" as the ability of a model, when faced with a question, to find the correct document or the correct source to answer it.
Hebbia's conclusions come from its own business scenarios, and a single-source experience account cannot replace systematic evaluation.
Hebbia's exact words were that Astra's leap in PowerPoint generation is "truly astonishing compared with other models."
Hebbia believes that evaluating the value of AI models in professional settings is shifting from generic benchmarks to the quality of specific deliverables.
The researcher read public materials from 12 teams running agents in production environments, including Codex, Claude Code, Cursor, Devin, Stripe's Minions, Manus, Agentforce, Shopify Sidekick, Sierra, Intercom Fin, Harvey, and Hebbia.
All Facts (7)
Hebbia says Astra can produce highly precise outputs and deliverables and follows the specifications set by users.
50%UnverifiedAfter testing OpenAI's Astra model, the Hebbia team saw a double-digit performance improvement on the citation source selection task compared with the other models they tested.
50%UnverifiedHebbia defines "citation source selection" as the ability of a model, when faced with a question, to find the correct document or the correct source to answer it.
50%UnverifiedHebbia's conclusions come from its own business scenarios, and a single-source experience account cannot replace systematic evaluation.
50%UnverifiedHebbia's exact words were that Astra's leap in PowerPoint generation is "truly astonishing compared with other models."
50%UnverifiedHebbia believes that evaluating the value of AI models in professional settings is shifting from generic benchmarks to the quality of specific deliverables.
50%UnverifiedThe researcher read public materials from 12 teams running agents in production environments, including Codex, Claude Code, Cursor, Devin, Stripe's Minions, Manus, Agentforce, Shopify Sidekick, Sierra, Intercom Fin, Harvey, and Hebbia.
50%