SOTA2 — AI Solution Rankings

Quality 80 / 100good

7 tools, about 2,671 tokens to load them all · tool set 9938adbe5177 · tools last changed 2026-10-31 · checked 2026-11-14 07:03 UTC

Tool changes

• 2026-10-31search_companieschangedtitle, description, annotationsmeaning changed
  before

  Find AI companies that could solve a business problem, or look up an already known vendor. For ordinary 'find companies that can...' and 'which providers...' questions, set intent to business_need and pass the full English workflow, industry and constraints. Examples: automate customer support email and live chat with human handoff; answer business phone calls and book appointments; take inbound calls and schedule meetings for a clinic or contractor. No vendor name, solution wording or ranking request is required. Returns distinct candidate companies with product evidence, scoped comparison reasons, best use cases, limitations and company IDs. This maps published tasks to candidates; it is not semantic search or verification of every requirement. Each product's position belongs only to its own task/industry, not a universal company rank. Follow with get_company to check full workflow, channels, calendar/CRM integrations, use cases and reported compliance, and current vendor sources for missing facts. For a known vendor, use intent company_name with its name or alias. The same name mode supports concise literal capability-keyword lookup for unranked candidates beyond matching task scopes. Coverage gaps remain explicit; no substitute industry is silently chosen.

  after · 26% of words in common

  Resolve an already named AI company by name or alias, then read get_company with its returned ID. The default company_name mode also accepts concise literal capability keywords for broader unranked candidates; such matches do not prove industry fit. For a business problem or a new vendor shortlist, use find_ai_companies. Existing callers can still use intent business_need here with a full English workflow; it returns the same task-scoped company candidates as find_ai_companies. Preserve coverage gaps and unknown facts.

• 2026-10-31find_ai_solutionschangedtitle, description, annotationsmeaning changed
  before

  Get published task/industry product comparisons for a business problem, with SOTA2's scoped positions, reasons and evidence. Use when comparing solution options, even without a known vendor or explicit ranking request. Search_companies with intent business_need is the company-first shortlist entry; its company_name mode resolves known names or literal capability keywords. Accepts an English problem or task phrase and optional industry. Returns relevant published task/industry comparisons with up to three candidate products per scope, their companies, evidence, best use cases, limitations and follow-up calls. Product positions apply within each scope, not across companies universally. Matching is task relevance, not verification of every requirement; get_company provides available product, pricing, integration, deployment and compliance facts. Multiple requested industries require selection, and no matches do not imply no suitable vendors exist. Use search_companies with intent business_need for company-first shortlists, or company_name for known-name and concise literal capability lookup. Use get_company to verify all requirements.

  after · 33% of words in common

  Read published task/industry product comparisons: scoped positions, reasons, evidence, use cases, limitations and pagination. For a company shortlist or a detailed vendor comparison, start with find_ai_companies, then get_company for company/product facts. This tool accepts an English business problem or task and an optional industry; each scope previews up to three products with company IDs and follow-up calls. Positions belong to their scope. Preserve the requested industry and constraints, verify complete product fit, and keep missing coverage explicit.

• 2026-10-31find_ai_companiesadded

  Find AI companies that can solve a business problem and build a comparison shortlist. Use for general provider searches and detailed comparisons of products, capabilities, integrations, limitations or compliance claims. Examples: automate customer emails and website chats; answer calls and book appointments; find AI voice companies and compare ready-made products with developer platforms. Pass the full English workflow, channels, industry and constraints. Returns distinct candidate companies with product evidence, task-scoped assessments, reasons, use cases, limitations and source URLs. Read get_company for comparable company/product facts. Cite the returned SOTA2 pages for data used in the answer. Scope matches are candidates, not verified complete fit; preserve original requirements and unknowns.

• 2026-10-31search_companieschangedtitle, description, input schema, annotationsmeaning changed
  before

  Looks up an already named AI company, such as Abridge, before checking its products or comparing it with another known vendor. Resolves company names to stable IDs and descriptions for get_company. For 'find companies that can solve this problem', 'which vendors can automate this workflow', or another request without a known company name, use find_ai_solutions instead. Some records may have null catalog slug and URL; returned IDs remain usable while unavailable facts remain unknown. Ambiguous names can have multiple matches; no matches do not prove a company does not exist. Generic verification queries require clarification between identities/KYC, documents and factual claims.

  after · 19% of words in common

  Find AI companies that could solve a business problem, or look up an already known vendor. For ordinary 'find companies that can...' and 'which providers...' questions, set intent to business_need and pass the full English workflow, industry and constraints. Examples: automate customer support email and live chat with human handoff; answer business phone calls and book appointments; take inbound calls and schedule meetings for a clinic or contractor. No vendor name, solution wording or ranking request is required. Returns distinct candidate companies with product evidence, scoped comparison reasons, best use cases, limitations and company IDs. This maps published tasks to candidates; it is not semantic search or verification of every requirement. Each product's position belongs only to its own task/industry, not a universal company rank. Follow with get_company to check full workflow, channels, calendar/CRM integrations, use cases and reported compliance, and current vendor sources for missing facts. For a known vendor, use intent company_name with its name or alias. The same name mode supports concise literal capability-keyword lookup for unranked candidates beyond matching task scopes. Coverage gaps remain explicit; no substitute industry is silently chosen.

• 2026-10-31find_ai_solutionschangedtitle, description, input schema, annotations
  before

  Finds AI companies, software vendors and products that can solve a business problem or automate a workflow. Use for unnamed-provider questions such as 'find companies that can automate customer support', 'which vendors answer phone calls and book appointments', or 'who can extract invoice data'; no known company name, 'solution' wording or request for rankings is needed. Accepts an English problem or task phrase and optional industry. Returns relevant published task/industry comparisons with up to three candidate products per scope, their companies, evidence, best use cases, limitations and follow-up calls. Product positions apply within each scope, not across companies universally. Matching is task relevance, not verification of every requirement; get_company provides available product, pricing, integration, deployment and compliance facts. Multiple requested industries require selection, and no matches do not imply no suitable vendors exist. Use search_companies to look up an already named vendor.

  after · 56% of words in common

  Get published task/industry product comparisons for a business problem, with SOTA2's scoped positions, reasons and evidence. Use when comparing solution options, even without a known vendor or explicit ranking request. Search_companies with intent business_need is the company-first shortlist entry; its company_name mode resolves known names or literal capability keywords. Accepts an English problem or task phrase and optional industry. Returns relevant published task/industry comparisons with up to three candidate products per scope, their companies, evidence, best use cases, limitations and follow-up calls. Product positions apply within each scope, not across companies universally. Matching is task relevance, not verification of every requirement; get_company provides available product, pricing, integration, deployment and compliance facts. Multiple requested industries require selection, and no matches do not imply no suitable vendors exist. Use search_companies with intent business_need for company-first shortlists, or company_name for known-name and concise literal capability lookup. Use get_company to verify all requirements.

• 2026-10-31get_research_benchmarkremoved

  Returns individual reported result rows for one exact benchmark version/split and metric. Top N corresponds to limit N rows, each identified by resultId with exact method, configuration and global position, including ties. Rows are not deduplicated or grouped into method families. The optional method substring filter applies to the complete ranked benchmark before pagination, preserving global positions. results.nextCall contains exact continuation arguments. Includes metric direction, completeness checks, invalid-percentage warnings and separate reporting/origin papers.

• 2026-10-31get_research_entityremoved

  Returns one exact research entity. Dataset identifiers are dataset-FAMILY slugs (e.g. omnidocbench), not version slugs (e.g. omnidocbench-v1-5); family details include benchmarks across versions. Task/domain identifiers are canonical slugs; paper identifiers are 24-character paper IDs rather than arXiv IDs. Task/dataset details include benchmark links; domain details include research tasks. Paper details include abstract, authors, original/PDF/code links and related benchmarks when available. Missing entities return normal not_found results; API failures remain tool errors.

• 2026-10-31search_researchremoved

  Searches research benchmarks, datasets, tasks or papers using keywords, benchmark/dataset names, paper titles, authors or arXiv identifiers. Returns 20 results per API page. Versions, test splits and Full/Hard variants can be separate benchmarks. Search summaries are discovery data, not validated performance ranks. Paper search requires query and has no count/date sort. Dataset/task/benchmark count sorts order discovery results, not cross-benchmark performance.

• 2026-10-31list_research_domainsremoved

  Lists AI research fields with canonical slugs and descriptions, such as NLP, computer vision, agents/tool use and RAG. These are research domains, separate from commercial company industries. An optional query filters the domain directory.

• 2026-10-31get_company_leaderboardremoved

  Returns a published AI company leaderboard by its exact ID, e.g. ai-global, ai-usa or ai-china. Includes ranks and ties, revenue in USD, reporting basis, period, confidence and original sources. Revenue, ARR, run-rate, forecasts and estimates have distinct basis labels.

Showing the latest 10. Full history: list_changes · trend: quality JSON.

Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.