| 2026-10-22 | parallel.ai/pricing | Run up to 5,000 requests per month for free; Search API latency 200 ms to 3 s |
| 2026-10-30 | parallel.ai/blog/free-tier-parallel | the USD 5 monthly credit goes to an organization with a credit card; one card activates one organization; usage past USD 5 is billed to the card; USD 5 covers up to 5,000 Search requests |
| 2026-10-30 | parallel.ai/pricing | up to USD 80 in credits at signup and USD 5 in free credits a month |
| 2026-10-30 | parallel.ai/blog/free-tier-parallel | a new credit arrives each month and unused balance expires at month end |
| 2026-10-30 | parallel.ai/pricing | up to 5,000 requests a month free; the same page lists 600 requests a minute for Search and 2,000 for Task |
| 2026-10-30 | docs.parallel.ai/integrations/mcp/quickstart | the hosted MCP server serves web_search, web_fetch and web_decisions free with no API key, at lower rate limits |
| 2026-10-30 | docs.parallel.ai/resources/rate-limits | the default Search quota is 600 requests a minute, counted per POST to /v1/search |
| 2026-11-01 | parallel.ai/pricing | the pricing page names the metered plan that carries the USD 5 monthly credit Pay as you go |
| 2026-11-01 | parallel.ai/articles/best-free-web-search-api | the vendor's comparison table states the hosted MCP server needs no account, up to USD 80 is earned at signup, and a card is needed only for the monthly credit |
| 2026-11-01 | platform.parallel.ai/signup | the platform sign-up page offers sign-up with an email address or with a Google account |
| 2026-11-01 | docs.parallel.ai/resources/faqs | monthly spend limits email the organization's admins and never block requests |
| 2026-11-01 | parallel.ai/blog/free-tier-parallel | the unused part of the monthly credit expires at the end of the month |
| 2026-11-01 | parallel.ai/pricing | the pricing page states up to 5,000 requests a month free |
| 2026-11-01 | docs.parallel.ai/resources/rate-limits | the default Search quota is 600 requests a minute, counted per POST to /v1/search |
| 2026-11-01 | docs.parallel.ai/resources/rate-limits | the default Task quota is 2,000 runs a minute |
| 2026-11-01 | docs.parallel.ai/integrations/mcp/quickstart | the hosted MCP server at mcp.parallel.ai/mcp serves web_search, web_fetch and web_decisions free with no API key, at lower rate limits |
| 2026-11-01 | parallel.ai/articles/best-free-web-search-mcp | the vendor states it does not publish the anonymous rate limit of the hosted MCP server |
| 2026-11-01 | probe: https://mcp.parallel.ai/mcp | a GET to the hosted MCP endpoint without credentials answers HTTP 405: the route takes POST |
| 2026-11-01 | probe: https://mcp.parallel.ai/mcp?login | a GET to the key-gated /mcp?login endpoint without credentials answers HTTP 401 |
| 2026-11-01 | probe: https://mcp.parallel.ai/mcp?login | the /mcp?login endpoint answers that it needs a Parallel API key or a sign-in |
| 2026-11-01 | parallel.ai/blog/free-tier-parallel | the USD 5 monthly credit goes to each eligible organization with a credit card |