VerbatimCV

A Textkernel (Sovren) alternative for AI agents

Choose VerbatimCV when an AI agent acts on the result and needs to know where each value came from: every preference carries its quote, missing facts are null, and hidden prompt injection is stripped and reported. Choose Textkernel's Tx Platform for job-title and skills normalisation, search and matching, and an established enterprise platform. Facts below come from Textkernel's own pages, read on 17 November 2026.

VerbatimCVTextkernel Tx Platform
Free start50 parses a month, no card500 free credits
Paid entryPaid plans not yet publishedProfessional from $99 a month (500–25,000 credits)
Cost per parseOne parse per request1 credit, +0.1 with the LLM Parser, +0.05 with OCR
Resume languagesText-layer CVs in most languages; OCR for 9 scripts29 languages
Uses an LLMYes, with layout repair, code-owned dates and quoted evidenceOptional LLM Parser add-on
Quote for each extracted preferenceYes (evidence)Not mentioned
Hidden text and prompt-injection handlingRemoved and reportedNot mentioned
MCP serverYesNot mentioned

About the LLM Parser

Textkernel's documentation says its LLM Parser "can generate information that is factually incorrect or misleading", omits some fields such as work-experience descriptions, accepts files up to about 3 MB, and falls back to the standard parser after errors or 60 seconds. VerbatimCV's answer to the same risk is a verbatim quote for every preference, so each value can be checked.

Where Textkernel is stronger

Normalisation of job titles and skills, geocoding, search and matching, more input formats (DOC, HTML, ODT, TXT) and a long enterprise track record.

Sources

"Not mentioned" means not stated on those pages on that date; the feature may exist elsewhere.

Try it

Get a free API key (50 parses a month, no card), then:

curl https://api.verbatimcv.com/v1/parse \
  -H "Authorization: Bearer $VERBATIMCV_API_KEY" \
  -F file=@cv.pdf

Agents can use the parse_resume tool instead: uvx verbatimcv-mcp. See the quickstart for Python, JavaScript and MCP setup.