Devtool Arena Changelog
All notable changes to the Devtool Arena leaderboard are documented here.
June 24, 2026
Synced new Browser CLI runs across Codex CLI and Claude Code CLI. Browserbase enters the Codex CLI top 5 at #4 with a score of 81, while Steel reaches #6 on Claude Code CLI; Firecrawl remains #1 on Codex CLI with 88.
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Browserbase | Browser | Cloud browser infrastructure for reliable web automation at scale | 81% | B | #4 of 55 | #1 of 4 |
| Steel | Browser | Cloud browser automation API and CLI for running browser sessions | 72% | C | #15 of 55 | #2 of 4 |
| Browser Use | Browser | AI-powered browser automation framework for web interactions | 69% | C | #18 of 55 | #3 of 4 |
| Anchor Browser | Browser | Headless browser automation API for web scraping and testing | 66% | C | #21 of 55 | #4 of 4 |
New company rankings above are based on the Codex CLI leaderboard.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 88% | B |
| 2 | WorkOS | Auth | 85% | B |
| 3 | LiveKit | Voice Infra | 82% | B |
| 4NEW | Browserbase | Browser | 81% | B |
| 5▼1 | Jina AI | Search | 78% | B |
| 6▼1 | Deepgram | Voice STT | 76% | B |
| 6▼1 | Resend | 76% | B | |
| 8▼1 | E2B | Sandboxes | 75% | B |
| 9▼1 | Stripe | Payment | 73% | C |
| 9▼1 | Datadog | Observability | 73% | C |
Top 10 based on the latest completed Codex CLI evaluation for each company.
Claude Code CLI Browser Results
Steel ranks #6 with a score of 79. Browserbase ranks #11 with 76, Anchor Browser ranks #15 with 73, and Browser Use ranks #16 with 72.
June 19, 2026
Synced the latest Claude Code API run for Reducto, lifting it into the top 10 at a #5 tie with a score of 84 and making it the leading Document Parsing result. Firecrawl remains #1 overall with a score of 93.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 93% | A |
| 2 | Box | Storage | 89% | B |
| 3 | ElevenLabs | Voice TTS | 86% | B |
| 3 | TinyFish | Search | 86% | B |
| 5 | Fireworks AI | Inference | 84% | B |
| 5 | Nimble | Search | 84% | B |
| 5 | Reducto | Document Parsing | 84% | B |
| 8▼1 | Chroma | Vector Databases | 83% | B |
| 8▼1 | Datadog | Observability | 83% | B |
| 8▼1 | Daytona | Sandboxes | 83% | B |
Top 10 based on the latest completed Claude Code API evaluation for each company.
Synced the latest Claude Code API Document Parsing runs for LlamaParse and Reducto. Reducto leads Document Parsing with a score of 76; LlamaParse improves to 74 and ties Unstructured, while Firecrawl remains #1 overall with 93.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 93% | A |
| 2 | Box | Storage | 89% | B |
| 3 | ElevenLabs | Voice TTS | 86% | B |
| 3 | TinyFish | Search | 86% | B |
| 5 | Fireworks AI | Inference | 84% | B |
| 5 | Nimble | Search | 84% | B |
| 7 | Chroma | Vector Databases | 83% | B |
| 7 | Datadog | Observability | 83% | B |
| 7 | Daytona | Sandboxes | 83% | B |
| 7 | OpenCageData | Geocoding | 83% | B |
Top 10 based on the latest completed Claude Code API evaluation for each company.
Synced the latest Claude Code API evaluations, including fresh Document Parsing runs for Reducto, Unstructured, and Extend.ai. Firecrawl remains #1 with a score of 93; Box is now #2 with 89, while ElevenLabs and TinyFish climb into the #3 tie.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 93% | A |
| 2 | Box | Storage | 89% | B |
| 3▲5 | ElevenLabs | Voice TTS | 86% | B |
| 3▲5 | TinyFish | Search | 86% | B |
| 5 | Fireworks AI | Inference | 84% | B |
| 5 | Nimble | Search | 84% | B |
| 7 | Chroma | Vector Databases | 83% | B |
| 7▲1 | Datadog | Observability | 83% | B |
| 7 | Anchor Browser | Browser | 83% | B |
| 7 | Daytona | Sandboxes | 83% | B |
Top 10 based on the latest completed Claude Code API evaluation for each company.
June 17, 2026
Synced the latest Codex CLI evaluations with 5 new companies across Code Review and Inference, including Pioneer AI, Semgrep, CodeRabbit, Greptile, and Qodo. Firecrawl remains #1 with a score of 88, while LiveKit, Jina AI, Deepgram, Resend, E2B, Datadog, and Stripe each move up one spot as Nimble drops out of the top 10 after its newer run.
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Pioneer AI | Inference | Inference platform for running and integrating AI models | 2% | D | #50 of 50 | #4 of 4 |
| Semgrep | Code Review | Static analysis and code security scanning CLI | 13% | D | #44 of 50 | #1 of 4 |
| CodeRabbit | Code Review | AI code review tooling for pull requests and repositories | 12% | D | #45 of 50 | #2 of 4 |
| Greptile | Code Review | Codebase intelligence and AI code review tooling | 12% | D | #45 of 50 | #2 of 4 |
| Qodo | Code Review | AI code quality and test generation tooling | 9% | D | #48 of 50 | #4 of 4 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 88% | B |
| 2 | WorkOS | Auth | 85% | B |
| 3▲1 | LiveKit | Voice Infra | 82% | B |
| 4▲1 | Jina AI | Search | 78% | B |
| 5▲1 | Deepgram | Voice STT | 76% | B |
| 5▲1 | Resend | 76% | B | |
| 7▲1 | E2B | Sandboxes | 75% | B |
| 8▲1 | Datadog | Observability | 73% | C |
| 8▲1 | Stripe | Payment | 73% | C |
| 10 | Agentmail | 72% | C |
Top 10 based on the latest completed Codex CLI evaluation for each company.
June 12, 2026
Synced the latest Claude Code MCP Document Parsing result for Extend.ai, improving its best score to 75 and keeping it #1 in Document Parsing. Exa remains #1 overall with a score of 93; the top 10 is unchanged.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Exa | Search | 93% | A |
| 2 | Tavily | Search | 90% | A |
| 3 | ElevenLabs | Voice TTS | 88% | B |
| 3 | Jina AI | Search | 88% | B |
| 5 | Firecrawl | Search | 87% | B |
| 6 | Stripe | Payment | 84% | B |
| 7 | Descope | Auth | 83% | B |
| 8 | Nimble | Search | 82% | B |
| 9 | Clerk | Auth | 81% | B |
| 9 | LanceDB | Vector Databases | 81% | B |
Top 10 based on the latest completed Claude Code MCP evaluation for each company.
Added Document Parsing to the Codex MCP leaderboard with Extend.ai entering the top 10 and Reducto at #12. Daytona takes #1 with 88, while Jina AI moves up to #3 (+5) and Firecrawl falls to #9 (-5).
New Categories
- Document Parsing — Document extraction and OCR-style parsing APIs for PDFs, images, and structured data
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Extend.ai | Document Parsing | Document processing API for extracting structured data from PDFs, images, and business documents | 78% | B | #10 of 70 | #1 of 2 |
| Reducto | Document Parsing | Document parsing API for converting PDFs and complex documents into structured data | 76% | B | #12 of 70 | #2 of 2 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1NEW | Daytona | Sandboxes | 88% | B |
| 2NEW | Exa | Search | 87% | B |
| 3▲5 | Jina AI | Search | 85% | B |
| 3 | Scrapfly | Webscraping | 85% | B |
| 5▼1 | Stripe | Payment | 83% | B |
| 6 | Descope | Auth | 81% | B |
| 6 | ElevenLabs | Voice TTS | 81% | B |
| 6▼2 | Tavily | Search | 81% | B |
| 9▼5 | Firecrawl | Search | 79% | B |
| 10NEW | Extend.ai | Document Parsing | 78% | B |
| 10 | Render | Cloud Hosting | 78% | B |
Top 10 based on the latest completed Codex MCP evaluation for each company.
June 2, 2026
Corrected Nimble on the Codex API and Codex CLI leaderboards after documentation/help text was counted as tool errors. Nimble now has 0 errors, ranking #3 on Codex CLI with a score of 84 and #18 on Codex API with a score of 77.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 88% | B |
| 2 | WorkOS | Auth | 85% | B |
| 3 | Nimble | Search | 84% | B |
| 4▼1 | LiveKit | Voice Infra | 82% | B |
| 5▼1 | Jina AI | Search | 78% | B |
| 6▼1 | Deepgram | Voice STT | 76% | B |
| 6▼1 | Resend | 76% | B | |
| 8▼1 | E2B | Sandboxes | 75% | B |
| 9▼1 | Datadog | Observability | 73% | C |
| 9▼1 | Stripe | Payment | 73% | C |
Top 10 based on the latest completed Codex CLI evaluation for each company. Codex API was also corrected for Nimble: #18 of 125 with a score of 77.
June 1, 2026
Expanded both leaderboards with 3 new categories (Browser, Document Parsing, Storage) and 12 new companies. Box leads the new Storage category with scores of 85 (Claude Code) and 83 (Codex). Vapi and Bolna join Voice Telephony. Diffbot added to Webscraping.
New Categories
- Browser — Headless browser automation and cloud browser APIs for web scraping and testing
- Document Parsing — Document extraction and parsing APIs for PDFs, images, and structured data
- Storage — Cloud storage integration APIs for file upload, sharing, and management
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Anchor Browser | Browser | Headless browser automation API for web scraping and testing | 77% | B | #22 of 118 | #1 of 3 |
| Browserbase | Browser | Cloud browser infrastructure for reliable web automation at scale | 72% | C | #30 of 118 | #2 of 3 |
| Browser Use | Browser | AI-powered browser automation framework for web interactions | 62% | C | #65 of 118 | #3 of 3 |
| Unstructured | Document Parsing | Open-source document preprocessing and extraction for LLM pipelines | 64% | C | #54 of 118 | #1 of 4 |
| Sensible | Document Parsing | Intelligent document extraction API for structured data from PDFs and images | 64% | C | #55 of 118 | #2 of 4 |
| Reducto | Document Parsing | Document intelligence API for extracting structured data from complex documents | 34% | D | #81 of 118 | #3 of 4 |
| LlamaParse | Document Parsing | LlamaIndex document parser for extracting text and tables from PDFs | 32% | D | #87 of 118 | #4 of 4 |
| Box | Storage | Cloud content management and file sharing platform with enterprise security | 85% | B | #11 of 118 | #1 of 2 |
| Dropbox | Storage | Cloud storage and file synchronization service with collaboration features | 70% | C | #35 of 118 | #2 of 2 |
| Vapi | Voice Telephony | Voice AI platform for building conversational voice agents and assistants | 60% | C | #74 of 118 | #3 of 8 |
| Bolna | Voice Telephony | Voice AI infrastructure for building telephony and voice agent applications | 37% | D | #80 of 118 | #5 of 8 |
| Diffbot | Webscraping | AI-powered web data extraction and knowledge graph API | 65% | C | #48 of 118 | #2 of 2 |
Scores by Platform
Claude Code API:
- Box: 85 | Anchor Browser: 77 | Browserbase: 72 | Dropbox: 70
- Diffbot: 65 | Sensible: 64 | Unstructured: 64 | Browser Use: 62
- Vapi: 60 | Bolna: 37 | Reducto: 34 | LlamaParse: 32
Codex API:
- Box: 83 | Dropbox: 76 | Browserbase: 73 | Diffbot: 72
- LlamaParse: 67 | Reducto: 66 | Browser Use: 65 | Anchor Browser: 65
- Vapi: 60 | Unstructured: 33 | Bolna: 30 | Sensible: 28
May 21, 2026
Re-ran the Claude Code leaderboard with updated evaluations. LiveKit returns to the top 10 at #7 with a score of 89, pushing the previous rank-7 group to #8. Jina AI achieves 88, joining a 6-way tie at #8 alongside Datadog, Cloudflare Workers, OpenCageData, ElevenLabs, and TinyFish. Modal improves to 78.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | You.com | Search | 93% | A |
| 1 | Exa | Search | 93% | A |
| 1 | Firecrawl | Search | 93% | A |
| 4 | Tavily | Search | 90% | A |
| 4 | Scrapfly | Webscraping | 90% | A |
| 4 | OpenRouter | Inference | 90% | A |
| 7 | LiveKit | Voice Infra | 89% | B |
| 8▼1 | Datadog | Observability | 88% | B |
| 8▼1 | Cloudflare Workers | Sandboxes | 88% | B |
| 8▼1 | OpenCageData | Geocoding | 88% | B |
| 8▼1 | ElevenLabs | Voice TTS | 88% | B |
| 8▼1 | TinyFish | Search | 88% | B |
| 8 | Jina AI | Search | 88% | B |
Top 10 based on the latest completed Claude Code evaluation for each company.
May 20, 2026
Re-ran the Codex CLI leaderboard with updated evaluations. Firecrawl climbs to #1 with a score of 88 (+2). WorkOS enters the top 3 at #2. LiveKit drops to #3 (-2) and Jina AI to #4 (-2). Deepgram moves up to #5 (+1).
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1▲2 | Firecrawl | Search | 88% | B |
| 2 | WorkOS | Auth | 85% | B |
| 3▼2 | LiveKit | Voice Infra | 82% | B |
| 4▼2 | Jina AI | Search | 78% | B |
| 5▲1 | Deepgram | Voice STT | 76% | B |
| 5▼1 | Resend | 76% | B | |
| 7▼3 | E2B | Sandboxes | 75% | B |
| 8 | Datadog | Observability | 73% | C |
| 8▼1 | Stripe | Payment | 73% | C |
| 10 | Render | Cloud Hosting | 72% | C |
Top 10 based on the latest completed Codex CLI evaluation for each company.
May 19, 2026
Expanded both the Claude Code and Codex leaderboards to 111 companies with 7 new categories (Geocoding, Verification, Unified API, CMS, Code Review, Proxy, Talent Directory) and 21 new companies. OpenCageData enters at #7 on Claude Code with a score of 88. Three-way tie for #1 on Claude Code: You.com, Exa, and Firecrawl all score 93.
New Categories
- Geocoding — Location lookup, address geocoding, and reverse geocoding APIs
- Verification — Identity verification, phone number validation, and fraud detection APIs
- Unified API — Aggregated APIs abstracting multiple underlying services behind one interface
- CMS — Headless content management and delivery APIs
- Code Review — Automated code analysis, review, and quality-gate APIs
- Proxy — Residential and datacenter proxy network APIs for web access
- Talent Directory — Developer and talent directory APIs for sourcing and profile data
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| OpenCageData | Geocoding | Forward and reverse geocoding API built on open data | 88% | B | #9 of 111 | #1 of 1 |
| LMNT | Voice TTS | Ultra-low-latency text-to-speech API for real-time voice applications | 81% | B | #28 of 111 | #2 of 5 |
| Rev AI | Voice STT | Asynchronous and real-time speech-to-text transcription API | 73% | C | #61 of 111 | #3 of 5 |
| Restate | Durable Workflow | Durable execution framework for distributed workflows and async services | 73% | C | #59 of 111 | #7 of 8 |
| Apideck | Unified API | Unified API layer aggregating CRM, HR, accounting, and other SaaS data | 72% | C | #62 of 111 | #1 of 1 |
| Agora | Voice Infra | Real-time voice, video, and live streaming infrastructure SDK | 69% | C | #71 of 111 | #2 of 5 |
| HitPay | Payment | Payment platform for Southeast Asian businesses | 64% | C | #77 of 111 | #8 of 11 |
| Prelude | Verification | Phone number verification and OTP API for user authentication | 59% | C | #82 of 111 | #2 of 2 |
| Surge | Verification | Identity and KYC verification API for onboarding flows | 59% | C | #81 of 111 | #1 of 2 |
| Nimble | Search | Web data extraction and SERP API for structured search results | 58% | C | #84 of 111 | #7 of 9 |
| Daily | Voice Infra | Real-time audio and video infrastructure API for live communication apps | 57% | C | #87 of 111 | #3 of 5 |
| BetterStack | Observability | Uptime monitoring, log management, and incident response platform | 61% | C | #78 of 111 | #2 of 2 |
| BlindPay | Stablecoin | Stablecoin payment rails for cross-border transfers | 44% | D | #96 of 111 | #5 of 6 |
| Upsun | Cloud Hosting | PaaS platform for deploying and scaling web applications | 34% | D | #100 of 111 | #5 of 5 |
| Ash | Sandboxes | Cloud sandbox environments for ephemeral dev and test workloads | 30% | D | #102 of 111 | #8 of 8 |
| Speechmatics | Voice STT | Enterprise-grade speech recognition API with broad language support | 21% | D | #106 of 111 | #5 of 5 |
| 100ms | Voice Infra | Real-time audio and video conferencing infrastructure SDK | 17% | D | #107 of 111 | #5 of 5 |
| Macroscope | Code Review | Automated code review and engineering analytics platform | 17% | D | #109 of 111 | #1 of 1 |
| Strapi | CMS | Open-source headless CMS with a REST and GraphQL content API | 17% | D | #110 of 111 | #1 of 1 |
| Massive | Proxy | Residential proxy network API for web data access at scale | 25% | D | #105 of 111 | #1 of 1 |
| Roster | Talent Directory | Developer talent directory and sourcing API | 8% | D | #111 of 111 | #1 of 1 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | You.com | Search | 93% | A |
| 1NEW | Exa | Search | 93% | A |
| 1▲4 | Firecrawl | Search | 93% | A |
| 4▼2 | Tavily | Search | 90% | A |
| 4NEW | Scrapfly | Webscraping | 90% | A |
| 4▲4 | OpenRouter | Inference | 90% | A |
| 7▼4 | Datadog | Observability | 88% | B |
| 7▼4 | Cloudflare Workers | Sandboxes | 88% | B |
| 7NEW | OpenCageData | Geocoding | 88% | B |
| 7NEW | ElevenLabs | Voice TTS | 88% | B |
| 7▲1 | TinyFish | Search | 88% | B |
Top 10 based on the latest completed Claude Code evaluation for each company.
May 16, 2026
Re-ran the Codex CLI leaderboard across most companies. LiveKit returns to #1 with 89; Firecrawl drops to #3 (-2). Added 3 new companies: Telnyx and Vonage (Voice Telephony) and Nylas (Email).
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Telnyx | Voice Telephony | Programmable voice, SMS, and fax API for real-time cloud communications | 52% | D | #18 of 39 | #1 of 3 |
| Vonage | Voice Telephony | Cloud communications platform for voice, SMS, and video APIs | 50% | D | #19 of 39 | #2 of 3 |
| Nylas | Email, calendar, and contacts API for embedding communication features | 27% | D | #29 of 39 | #3 of 3 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | LiveKit | Voice Infra | 89% | B |
| 2 | Jina AI | Search | 88% | B |
| 3▼2 | Firecrawl | Search | 84% | B |
| 4▲1 | Resend | 82% | B | |
| 4▲1 | E2B | Sandboxes | 82% | B |
| 6 | Deepgram | Voice STT | 81% | B |
| 7 | Stripe | Payment | 80% | B |
| 8 | Datadog | Observability | 77% | B |
| 9 | Sprites | Sandboxes | 75% | B |
| 10 | Agentmail | 73% | C |
Top 10 based on the latest completed Codex CLI evaluation for each company.
May 15, 2026
Tavily climbs to #2 on the Claude Code leaderboard with a new top score of 90, up from 86 (+2 ranks). Nebius Token Factory joins the Inference category. TinyFish also adds a Codex run at score 82.
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Nebius Token Factory | Inference | Fast LLM inference API with flexible model routing and token-based pricing | 74% | C | #40 of 94 | #5 of 9 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | You.com | Search | 93% | A |
| 2▲2 | Tavily | Search | 90% | A |
| 3▼1 | Datadog | Observability | 88% | B |
| 3▼1 | Cloudflare Workers | Sandboxes | 88% | B |
| 5▼1 | Firecrawl | Search | 86% | B |
| 5▼1 | Vercel | Sandboxes | 86% | B |
| 5▼1 | Stripe | Payment | 86% | B |
| 8 | OpenRouter | Inference | 84% | B |
| 8 | TinyFish | Search | 84% | B |
| 8 | Jina AI | Search | 84% | B |
Top 10 based on the latest completed Claude Code evaluation for each company.
May 13, 2026
Expanded the Codex MCP leaderboard to 93 companies. Added 5 new companies including TinyFish, Convex, and Anchor Browser. You.com enters at #1 with a score of 93 — the highest MCP score yet — while Firecrawl climbs to #4 (+2).
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| TinyFish | Search | AI-powered search and discovery API | 84% | B | #10 of 93 | #3 of 8 |
| Convex | Database | Reactive backend-as-a-service with real-time database and serverless functions | 73% | C | #46 of 93 | #1 of 1 |
| Anchor Browser | Browser Automation | Headless browser automation API for web scraping and testing | 67% | C | #58 of 93 | #1 of 1 |
| TABStack | Search | Developer search and code intelligence API | 54% | D | #77 of 93 | #5 of 8 |
| Freestyle | Sandboxes | Cloud sandbox environments for running untrusted code | 53% | D | #78 of 93 | #7 of 7 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1NEW | You.com | Search | 93% | A |
| 2NEW | Datadog | Observability | 88% | B |
| 2NEW | Cloudflare Workers | Sandboxes | 88% | B |
| 4NEW | Vercel | Sandboxes | 86% | B |
| 4▲2 | Firecrawl | Search | 86% | B |
| 4▼2 | Stripe | Payment | 86% | B |
| 4▼1 | Tavily | Search | 86% | B |
| 8NEW | OpenRouter | Inference | 84% | B |
| 8▼7 | Jina AI | Search | 84% | B |
| 8NEW | TinyFish | Search | 84% | B |
Top 10 based on the latest completed MCP evaluation for each company.
May 9, 2026
Updated the Codex MCP leaderboard with gpt-5.5 runs across 68 companies. Jina AI holds #1 with 85; Stripe enters at #2 with 83. Added 8 new companies including Telnyx, Twilio, Resemble AI, Pipecat, and Fireworks AI.
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Telnyx | Voice Telephony | Programmable voice, messaging, and SIP trunking API | 66% | C | #13 of 68 | #1 of 7 |
| Plivo | Voice Telephony | Voice and SMS API platform for building communications apps | 59% | C | #19 of 68 | #2 of 7 |
| Twilio | Voice Telephony | Cloud communications platform for voice, SMS, and video | 57% | C | #22 of 68 | #3 of 7 |
| Resemble AI | Voice TTS | AI voice cloning and text-to-speech synthesis API | 47% | D | #27 of 68 | #3 of 3 |
| Sinch | Voice Telephony | Cloud communications API for voice, SMS, and messaging | 10% | D | #52 of 68 | #4 of 7 |
| Vonage | Voice Telephony | Programmable communications API for voice and messaging | 10% | D | #52 of 68 | #4 of 7 |
| Pipecat | Voice Infra | Open-source framework for real-time voice AI pipelines | 4% | D | #66 of 68 | #2 of 2 |
| Fireworks AI | Inference | Fast inference API for open-source LLMs and image models | 4% | D | #66 of 68 | #5 of 5 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Jina AI | Search | 85% | B |
| 2 | Stripe | Payment | 83% | B |
| 3 | ElevenLabs | Voice TTS | 81% | B |
| 3 | Tavily | Search | 81% | B |
| 5 | Scrapfly | Webscraping | 80% | B |
| 6▼4 | Firecrawl | Search | 78% | B |
| 6 | Render | Cloud Hosting | 78% | B |
| 8 | Descope | Auth | 77% | B |
| 9▼6 | Agentmail | 74% | C | |
| 10 | Qdrant | Vector Databases | 72% | C |
Top 10 based on the latest completed MCP evaluation for each company.
May 8, 2026
Launched the Codex MCP leaderboard with 53 companies scored. Added Voice Telephony as a new category with Bandwidth and Infobip. Jina AI leads at #1 with a score of 81; Firecrawl and Agentmail round out the top 3.
New Categories
- Voice Telephony — Telephony and programmable voice call APIs
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Nylas | Email and calendar API for reading, sending, and managing messages | 19% | D | #9 of 53 | #3 of 3 | |
| Coinbase Payments | Stablecoin | Stablecoin and crypto payments API from Coinbase | 10% | D | #37 of 53 | #1 of 2 |
| Fireblocks | Stablecoin | Digital asset custody and stablecoin transfer infrastructure | 10% | D | #39 of 53 | #2 of 2 |
| Bandwidth | Voice Telephony | Programmable voice and messaging carrier API | 9% | D | #43 of 53 | #1 of 2 |
| Infobip | Voice Telephony | Omnichannel communications platform with voice call and SMS APIs | 4% | D | #53 of 53 | #2 of 2 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1NEW | Jina AI | Search | 81% | B |
| 2NEW | Firecrawl | Search | 71% | C |
| 3NEW | Agentmail | 68% | C | |
| 4NEW | Resend | 65% | C | |
| 5NEW | Clerk | Auth | 63% | C |
| 6NEW | Cartesia | Voice TTS | 61% | C |
| 7NEW | Chroma | Vector Databases | 48% | D |
| 8NEW | Cerebras | Inference | 34% | D |
| 9NEW | Nylas | 19% | D | |
| 10NEW | Qdrant | Vector Databases | 13% | D |
Top 10 based on the latest completed Codex MCP evaluation for each company.
April 30, 2026
Updated the Claude Code API, Claude Code CLI, Codex API, and Codex CLI leaderboards. This snapshot reflects the latest completed runs on this date, with 108 companies currently listed on the leaderboard and 954 completed evaluations.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | ElevenLabs | Voice TTS | 92% | A |
| 1 | Firecrawl | Search | 92% | A |
| 3 | Auth0 | Auth | 91% | A |
| 4 | Tavily | Search | 90% | A |
| 5 | LiveKit | Voice Infra | 89% | B |
| 6 | Datadog | Observability | 88% | B |
| 7 | Stripe | Payment | 86% | B |
| 8 | Jina AI | Search | 85% | B |
| 9 | OpenRouter | Inference | 84% | B |
| 10 | Deepgram | Voice STT | 83% | B |
Top 10 based on the latest completed evaluation for each company across all platforms.
April 29, 2026
Updated the Codex CLI leaderboard. This snapshot reflects the latest completed Codex CLI run on this date, with 88 companies currently listed on the leaderboard and 216 completed Codex CLI evaluations.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 86% | B |
| 2 | Jina AI | Search | 85% | B |
| 3 | ElevenLabs | Voice TTS | 84% | B |
| 3 | Tavily | Search | 84% | B |
| 5 | E2B | Sandboxes | 82% | B |
| 5 | Resend | 82% | B | |
| 7 | Stripe | Payment | 80% | B |
| 8 | AssemblyAI | Voice STT | 79% | B |
| 8 | Mollie | Payment | 79% | B |
| 8 | WorkOS | Auth | 79% | B |
Top 10 based on the latest completed Codex CLI evaluation for each company.
April 26, 2026
Added the Codex CLI leaderboard. This snapshot reflects the latest completed Codex CLI end-to-end run on this date, with 74 companies currently listed on the leaderboard and 33 completed Codex CLI evaluations.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | LiveKit | Voice Infra | 89% | B |
| 2 | Firecrawl | Search | 86% | B |
| 3 | Jina AI | Search | 85% | B |
| 4 | Tavily | Search | 84% | B |
| 5 | E2B | Sandboxes | 82% | B |
| 5 | Resend | 82% | B | |
| 7 | Stripe | Payment | 80% | B |
| 8 | WorkOS | Auth | 79% | B |
| 9 | Datadog | Observability | 77% | B |
| 10 | Sprites | Sandboxes | 75% | B |
Top 10 based on the latest completed Codex CLI end-to-end evaluation for each company.
April 24, 2026
Added the Codex API leaderboard. This snapshot reflects the latest completed Codex API run on this date, with 74 companies run.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | ElevenLabs | Voice TTS | 92% | A |
| 1 | Firecrawl | Search | 92% | A |
| 3 | Tavily | Search | 90% | A |
| 4 | Datadog | Observability | 88% | B |
| 5 | Stripe | Payment | 86% | B |
| 6 | OpenRouter | Inference | 84% | B |
| 7 | Deepgram | Voice STT | 83% | B |
| 8 | Cartesia | Voice TTS | 79% | B |
| 8 | Daytona | Sandboxes | 79% | B |
| 8 | Mollie | Payment | 79% | B |
Top 10 based on the latest completed Codex API evaluation for each company.
April 15, 2026
Added the MCP leaderboard. This snapshot reflects the latest completed MCP run on this date, with 74 companies run on the leaderboard and 50 companies currently tracked with MCP metadata in the registry.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Chroma | Vector Databases | 78% | B |
| 2 | Agentmail | 77% | B | |
| 2 | Qdrant | Vector Databases | 77% | B |
| 2 | Tavily | Search | 77% | B |
| 5 | Firecrawl | Search | 76% | B |
| 5 | Jina AI | Search | 76% | B |
| 7 | ElevenLabs | Voice TTS | 75% | B |
| 7 | MeetGeek | Meeting Bot | 75% | B |
| 9 | Deepgram | Voice STT | 74% | C |
| 9 | Descope | Auth | 74% | C |
Top 10 based on the latest completed MCP evaluation for each company.
April 9, 2026
Added the Claude Code CLI leaderboard. This snapshot reflects the latest completed Claude Code CLI end-to-end run on this date, with 74 companies run on the leaderboard and 35 companies currently tracked with CLI metadata in the registry.
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | LiveKit | Voice Infra | 83% | B |
| 2 | Firecrawl | Search | 80% | B |
| 3 | Jina AI | Search | 79% | B |
| 3 | Resend | 79% | B | |
| 5 | Daytona | Sandboxes | 78% | B |
| 5 | Tavily | Search | 78% | B |
| 7 | Vercel | Sandboxes | 77% | B |
| 8 | Sprites | Sandboxes | 74% | C |
| 8 | Stripe | Payment | 74% | C |
| 8 | WorkOS | Auth | 74% | C |
Top 10 based on the latest completed Claude Code CLI end-to-end evaluation for each company.
April 5, 2026
Added 2 new categories to cover APIs that didn't fit existing buckets, along with Datadog and Agentmail.
New Categories
- Observability — Monitoring, metrics, and event tracking APIs
- Email — Programmatic email inbox and messaging APIs
New Companies
| Company | Category | Description | Score | Grade | Overall Rank | Category Rank |
|---|---|---|---|---|---|---|
| Datadog | Observability | Python SDK for submitting custom metrics and events | 83% | B | #3 of 72 | #1 of 1 |
| Agentmail | Programmatic email inboxes, send/receive, and AI agent email | 75% | B | #17 of 72 | #1 of 1 |
Top 10 Overall
| Rank | Company | Category | Score | Grade |
|---|---|---|---|---|
| 1 | Firecrawl | Search | 86% | B |
| 2 | Jina AI | Search | 84% | B |
| 3NEW | Datadog | Observability | 83% | B |
| 4▼1 | Auth0 | Auth | 82% | B |
| 4▼1 | Prefect | Durable Workflow | 82% | B |
| 4▼1 | Stripe | Payment | 82% | B |
| 7▼1 | Tavily | Search | 81% | B |
| 7▼1 | PayPal | Payment | 81% | B |
| 9▼1 | You.com | Search | 80% | B |
| 9▼1 | ElevenLabs | Voice TTS | 80% | B |
March 2026
Launched the Devtool Arena leaderboard with 72 API companies scored.
Initial Launch
- Inference — Groq, TogetherAI, Perplexity, OpenRouter, Cerebras, Fireworks AI, DeepInfra, SambaNova
- Auth — Clerk, Auth0, Stytch, Scalekit, WorkOS, Descope
- Payment — Stripe, PayPal, Square, Adyen, Razorpay, Mollie, Paddle, Chargebee, LemonSqueezy, Airwallex
- Search — Exa, Tavily, You.com, Brave Search, Firecrawl, Jina AI
- Voice STT — Deepgram, AssemblyAI, OpenAI Whisper
- Voice TTS — ElevenLabs, Rime
- Voice Infra — LiveKit
- Sandboxes — Vercel, Cloudflare Workers, E2B, Daytona, Modal, Fly.io
- Vector Databases — Chroma, Pinecone, Weaviate, Qdrant, LanceDB, Zilliz Cloud
- Stablecoin — Stripe (Crypto), Coinbase Payments, Circle, Fireblocks, DFNS, Triple-A
- Meeting Bot — Recall.ai, Meeting BaaS, MeetGeek, Meetstream, CueMeet
- Durable Workflow — Temporal, Prefect, Cadence, Camunda, Netflix Conductor, AWS Step Functions, Akka
- Cloud Hosting — Vercel, Cloudflare, Netlify, Railway, Render, DigitalOcean
Join the mailing list
Get updates on leaderboard changes, new benchmark releases, and product announcements.