Devtool MCP Arena
Benchmarking developer tools evaluated by 
Claude Code across auth, inference, voice, and more.
SapientClickTap any row to explore the full eval.
Get your API graded →
| Rank | Tool | Score | Grade | Task | Eval | Discovery | Cost | Calls | Errors | Time | |
|---|---|---|---|---|---|---|---|---|---|---|---|
1🥇 | Tavily | 86 | B | 100 | 55 | $0.00 | 3 | 0 | 53s | View report | |
2🥈 | Chroma | 78 | B | 74 | 89 | $0.16 | 10 | 1 | 4m 9s | View report | |
3🥉 | Qdrant | 77 | B | 73 | 87 | $0.13 | 10 | 1 | 6m 25s | View report | |
3🥉 | Agentmail | 77 | B | 74 | 84 | $0.14 | 8 | 1 | 4m 57s | View report | |
5 | Firecrawl | 76 | B | 73 | 85 | $0.16 | 7 | 2 | 7m 19s | View report | |
5 | Jina AI | 76 | B | 74 | 82 | $0.13 | 6 | 1 | 4m 21s | View report | |
7 | MeetGeek | 75 | B | 71 | 86 | $0.11 | 7 | 1 | 5m 5s | View report | |
7 | ElevenLabs | 75 | B | 73 | 80 | $0.16 | 9 | 1 | 9m 41s | View report | |
9 | Descope | 74 | C | 71 | 80 | $0.21 | 10 | 1 | 3m 29s | View report | |
9 | Stripe | 74 | C | 73 | 77 | $0.10 | 7 | 1 | 6m 57s | View report | |
9 | Netlify | 74 | C | 74 | 75 | $0.18 | 9 | 2 | 3m 29s | View report | |
12 | Daytona | 73 | C | 67 | 88 | $0.20 | 18 | 2 | 3m 47s | View report | |
13 | Clerk | 72 | C | 68 | 84 | $0.31 | 15 | 1 | 8m 43s | View report | |
13 | Resend | 72 | C | 74 | 69 | $0.13 | 8 | 1 | 4m 2s | View report | |
15 | Pinecone | 71 | C | 66 | 83 | $0.25 | 16 | 3 | 5m 32s | View report | |
16 | Render | 70 | C | 70 | 71 | $0.16 | 13 | 1 | 5m 52s | View report | |
17 | Cartesia | 69 | C | 68 | 73 | $0.19 | 13 | 1 | 4m 48s | View report | |
18 | PayPal | 68 | C | 73 | 57 | $0.16 | 8 | 1 | 9m 53s | View report | |
18 | Recall.ai | 68 | C | 65 | 75 | $0.35 | 22 | 5 | 8m 45s | View report | |
20 | Zilliz Cloud | 62 | C | 66 | 55 | $0.42 | 16 | 5 | 3m 23s | View report | |
21 | Exa | 59 | C | 62 | 55 | $0.00 | 4 | 0 | 1m 23s | View report | |
22 | You.com | 55 | C | 40 | 91 | $0.12 | 6 | 1 | 4m 50s | View report | |
23 | Plivo | 45 | D | 31 | 79 | $0.32 | 19 | 8 | 3m 29s | View report | |
24 | LiveKit | 42 | D | 28 | 74 | $0.38 | 13 | 3 | 11m 47s | View report | |
25 | E2B | 40 | D | 34 | 55 | $0.00 | 8 | 0 | 1m 29s | View report | |
26 | Telnyx | 38 | D | 29 | 61 | $0.41 | 23 | 8 | 5m 46s | View report | |
27 | Prefect | 34 | D | 11 | 89 | $0.25 | 20 | 10 | 6m 31s | View report | |
28 | Cerebras | 25 | D | 6 | 71 | $0.52 | 30 | 11 | 5m 26s | View report | |
29 | Resemble AI | 24 | D | 6 | 68 | $0.74 | 26 | 5 | 4m 27s | View report | |
30 | LanceDB | 23 | D | 20 | 30 | $0.25 | 10 | 0 | 6m 39s | View report | |
31 | Razorpay | 22 | D | 13 | 45 | $0.70 | 36 | 0 | 8m 10s | View report | |
32 | Paddle | 20 | D | 6 | 55 | $0.64 | 28 | 13 | 5m 31s | View report | |
33 | Deepgram | 17 | D | 1 | 55 | $1.00 | 63 | 8 | 9m 42s | View report | |
34 | Datadog | 14 | D | 2 | 45 | $0.79 | 31 | 5 | 11m 49s | View report | |
34 | Meeting BaaS | 14 | D | 1 | 45 | $1.33 | 54 | 7 | 22m 21s | View report | |
36 | Scalekit | 12 | D | 2 | 38 | $0.94 | 31 | 1 | 17m 55s | View report | |
37 | Groq | 11 | D | 1 | 35 | $1.20 | 57 | 9 | 12m 48s | View report | |
38 | Auth0 | 10 | D | 0 | 34 | $0.66 | 27 | 8 | 4m 21s | View report | |
38 | Coinbase Payments | 10 | D | 0 | 35 | $1.54 | 79 | 8 | 39m 5s | View report | |
38 | Sinch Missing MCP auth env vars: SINCH_APPLICATION_KEY, SINCH_APPLICATION_SECRET | — | — | — | — | — | — | — | — | — | View report |
38 | Vonage Missing MCP auth env vars: VONAGE_APPLICATION_ID, VONAGE_PRIVATE_KEY | — | — | — | — | — | — | — | — | — | View report |
42 | Brave Search API key requires a paid plan | — | — | — | — | — | — | — | — | — | View report |
42 | Airwallex API key requires official certification | — | — | — | — | — | — | — | — | — | View report |
42 | Netflix Conductor Open-source / self-hosted — no hosted API key available | — | — | — | — | — | — | — | — | — | View report |
42 | Fireblocks No MCP server available | — | — | — | — | — | — | — | — | — | View report |
46 | Stytch MCP server requires browser-based OAuth (cannot automate) | — | — | — | — | — | — | — | — | — | View report |
47 | Railway No MCP server available | — | — | — | — | — | — | — | — | — | View report |
48 | Fireworks AI No MCP server package available | — | — | — | — | — | — | — | — | — | View report |
48 | Pipecat Open-source / self-hosted — no hosted API key available | — | — | — | — | — | — | — | — | — | View report |
50 | WorkOS No MCP server package available | — | — | — | — | — | — | — | — | — | View report |
51 | Perplexity API key requires a paid plan | — | — | — | — | — | — | — | — | — | View report |
51 | Cloudflare Workers No MCP server available | — | — | — | — | — | — | — | — | — | View report |
51 | TogetherAI API key requires a paid plan | — | — | — | — | — | — | — | — | — | View report |
51 | Adyen API key requires official certification | — | — | — | — | — | — | — | — | — | View report |
51 | Weaviate No MCP server available | — | — | — | — | — | — | — | — | — | View report |
51 | Vercel No MCP server available | — | — | — | — | — | — | — | — | — | View report |
51 | DigitalOcean Signup requires a credit card | — | — | — | — | — | — | — | — | — | View report |
51 | Square No MCP server available | — | — | — | — | — | — | — | — | — | View report |
Scores calculated from end-to-end MCP evaluations by AI coding agents.
Frequently asked questions
Join the mailing list
Get updates on leaderboard changes, new benchmark releases, and product announcements.