Codex vs Nansen x402 API

Codex 85

Hosted lines. Status page at status.codex.io (incident.io) with API, Dashboard and 25 network components and a history page (20). The history from 8 August to 6 November lists an elevated-latency incident on 11 August (balances queries, 42 minutes, labelled degraded performance), a 5xx spike the same day, degraded filterTokens on 12 August, and network delays on Robinhood, BNB, Cronos, Fogo and Solana pricing in late August and September, with none in October or November. All read as partial or degraded, so minor only (20). Per-plan rate limits (5 a second free, 300 on Growth with a 500 burst) and per-request maxima are published (15). 429 TOO_MANY_REQUESTS carries retryAfterSeconds and Retry-After, OVER_CAPACITY says to wait 30 seconds, and the errors page says which codes to retry; writes are limited to webhook and token mutations (15). The pricing page shows '99.9% Uptime SLA' as a headline figure and custom SLAs on Enterprise, with no SLA document found (5 of 10). The API is generally available (10). 20+20+15+15+5+10 = 85.

Nansen x402 API 45

We found no status page. status.nansen.ai still doesn't resolve on 1 November, and neither the API docs, llms.txt nor the site link one (0). No readable incident history (5). Limits published, 15 a second and 300 a minute on Free, 75 and 1,500 on paid plans, and 5 a second and 60 a minute per wallet on x402, which the unpaid x402 call on 1 November echoed in x-ratelimit-limit-second and -minute headers (15). 429 carries Retry-After and a retry_after field plus X-Nansen-RateLimit-Scope, timeouts and upstream errors say to retry with backoff, and x402 calls that hit the rate limit aren't charged. Every call is a read (15). No SLA found (0). We found no beta or preview label on the x402 route (10). 0 + 5 + 15 + 15 + 0 + 10 = 45.

Codex 80

MPP (Machine Payments Protocol) on every query endpoint, paid in USDC on Tempo; subscriptions and webhooks are not covered, and it is not x402 (35 of 40). $0.001 a query through MPP, compute-unit costs per operation, and Growth at $120 per million compute units, all public (20). The Almost free plan's 10,000 compute units need a one-time $1 activation fee by card or 1 USDC, which the terms say confirms a payment method (5 of 20). An agent can pay per query with no account or key (20). 35+20+5+20 = 80.

Nansen x402 API 95

x402 on Nansen's own api.nansen.ai for every Pro endpoint except labels, 17 at $0.01 and the Premium and Smart Money endpoints at $0.05. An unpaid call to current-balance on 1 November answered 402 with an x402 v2 PAYMENT-REQUIRED header asking $0.01 in eight ways, USDC on Base, Monad and Solana, USD₮0 on X Layer and four stablecoins on BNB Chain. The docs still name only Base, Solana and Monad. MPP on Tempo is documented for the same endpoints. Labels are excluded, and the 200 to 750-credit agent endpoints don't appear in the x402 tiers (35). Per-call prices published without login (20). Free plan with 100 trial credits and a daily top-up to 10, with no payment required (20). The docs call x402 "fully permissionless", a wallet with USDC is enough (20). 35 + 20 + 20 + 20 = 95.

Codex 59

Closed service under terms that name Codex Data Inc., last updated 29 October 2026; the SDK is MIT (15 of 30). The privacy policy (effective 29 September 2026) is general, keeps data 'only for as long as is necessary' with no periods, transfers data to the United States, and links no DPA or subprocessor list (10 of 30). The changelog dates deprecations, such as NFT endpoints removed on 31 March 2026 and the transactions fields deprecated on 2 November, with no written deprecation policy or notice period (12 of 20). The terms name Mixpanel, Rollbar and Sentry and the privacy policy says data is processed in the United States; no fuller subprocessor list found (8 of 20). 15+10+12+8 = 45.

Nansen x402 API 51

Closed service with published terms (updated 10 August 2026), API terms and a master services agreement (15). The privacy policy, re-read on 1 November, says API usage patterns, query parameters, IP addresses and timestamps are collected, keeps data "only for as long as it is necessary" with no period, and names the Nansen group rather than a legal entity (12). Deprecations come with dates. The address to wallet_address rename on 24 September 2026 keeps the old field until at least 23 December 2026 with a Warning: 299 header, and the 28 October perp-positions funding fields stay until at least 26 January 2027 (18). Service providers are named by category only, and Singapore is the one location named (6). 15 + 12 + 18 + 6 = 51.

Codex 91

The full GraphQL schema, 18,159 lines, is in the MIT SDK repository at src/resources/schema.graphql, updated on 5 November 2026, and introspection is free (25). llms.txt, llms-full.txt and a Markdown twin of every page (10). Reference pages give a one-line purpose per operation, and the API reference intro, the queries-versus-subscriptions page and llms.txt say which operation to use for which job and to batch rather than loop (13 of 20). GraphQL types with enums, input objects and non-null markers, and documented maxima per argument (14 of 15). Examples on every recipe and reference page, and an error table with code, HTTP status, meaning and retry advice (14 of 15). Weekly dated changelog from 26 August 2024 to 2 November 2026 with RSS and dated deprecations (15). 25+10+13+14+14+15 = 91.

Nansen x402 API 84

Each endpoint page embeds an OpenAPI 3.1 definition. We found no single downloadable spec (22). llms.txt at docs.nansen.ai, Markdown copies of every page and a ?ask= query on the docs (10). Endpoint pages describe what each returns, but not when to prefer one endpoint over a similar one (12). Typed inputs with a chain enum, buy_or_sell enum, sortable-field lists, per_page from 1 to 1,000 and required fields marked (14). An error catalogue with stable codes, request_id, doc_url and param. The who-bought-sold page has no worked request or response example (11). Dated changelog with one page per change (15).

Codex 79

GraphQL field selection, so a response holds only the fields asked for, and usageData reports the cost of a request (23 of 25). limit on filter queries up to 200, rankings and 100+ filter signals on filterTokens; some list arguments drop extra inputs silently rather than failing (18 of 20). Coded errors an agent can branch on, but most arrive as HTTP 200 in an errors array, and an inline input type mismatch returns only a message (16 of 20). Queries are reads and identical requests cost the same; errors are not billed except NOT_FOUND. No idempotency key on the webhook and token mutations (14 of 20). Few required arguments beyond networkId. One official SDK, TypeScript (8 of 15). 23+18+16+14+8 = 79.

Nansen x402 API 86

Pagination defaults to 10 rows (per_page up to 1,000) and filters on labels, volumes and dates keep responses small (20). page, per_page, order_by and range filters on the endpoints we read (20). Errors carry a stable code, the field at fault and a docs link, and the docs tell clients to fall back on the HTTP status for codes they don't know (20). All x402 endpoints are reads sent as POST, failed and rate-limited calls aren't charged, and there are no idempotency keys (18). Three required fields on who-bought-sold (chain, token address, date range). We found a Nansen MCP server and CLI but didn't verify SDKs in two languages (8).

Codex 80

Closed service. Newest changelog entry 2 November 2026 (30). Weekly entries from 3 August to 2 November, and SDK releases 2.5.10 to 2.5.20 between 28 October and 5 November (20). Public changelog, a Discord, a support address and a status page with a report link; GitHub issue replies were not sampled (10 of 15). Current official TypeScript SDK, no other languages (12 of 15). The SDK repository runs lint and test on pull requests and a release workflow (8 of 10). 30+20+10+12+8 = 80.

Nansen x402 API 75

Changelog entries on 30 October 2026, the newest of them a labelled breaking change to who-bought-sold (30). 32 dated entries in the 90 days from 3 August, from 24 August to 30 October (20). Public changelog and docs that answer questions through ?ask=. We didn't test support (10). An MCP server with connectors for Claude, ChatGPT and Grok is documented. We didn't verify official SDKs (10). No public package to check (5). 30 + 20 + 10 + 10 + 5 = 75.