CometAPI Pricing 2026

One prepaid API key over Veo, Kling, FLUX and ElevenLabs — prices shown upfront, failed jobs refunded

Pricing: Prepaid pay-as-you-go with no subscriptions and no minimums. You load any amount by card into a wallet, and each generation draws its exact listed price from the balance; the vendor states the rate shown is the rate charged, with no markup at checkout, and that funds do not expire. Per-model rates for every video, image and voice model are published on the pricing page, which the vendor describes as below list price, and the worked example given is an 8-second cinematic Veo 3 video at $0.40. Failed generations are automatically refunded to the balance, so only delivered output is billed. Signup is free and a playground is available for evaluation before loading funds. Support is by email at support@prizmkey.com.

Open-source OpenAI-compatible gateway routing one API to 200+ models across 40+ providers

Pricing: Free is $0 forever with no credit card, giving access to all 200+ models across 40+ providers, 3 rate-limited free models, bring-your-own-keys, prompt caching, budgets and spend controls, auto-routing and vendor selection, and 30 days of activity-log retention. The platform fee is 5% on credit usage. Enterprise is custom-priced with a 30-day trial, full control plus BYOK, custom routing rules, custom free-model limits, unlimited log retention, team management, advanced analytics and admin controls. Data retention is metered separately: metadata is free, full payload storage is $0.01 per 1M tokens.

One API key for 180+ AI models at 30–70% under official pricing, drop-in compatible with existing SDKs

Pricing: Usage-based credits with no subscription: the published base rate is 275,000 credits for $1,250, roughly $0.004545 per credit, and every model lists its per-run credit cost (for example, Nano Banana 2 text-to-image at 1K resolution is 3 credits, about $0.0136 per run). New accounts get 50 free credits. Billing is per successful call. The headline claim is 30–70% below official developer pricing for the same models.

Cloud desktops for AI agents — full Linux or Windows machines in under 500ms via API

Pricing: Three published monthly tiers, each bundling a small AI credit allowance and sold with monthly capacity add-ons. Hacker is $29 per month for solo builders and personal agents: 1 computer, 8 GB RAM, a 40 GB storage pool, 1 team member, Linux, provided templates, community support, the real-time collaborative canvas and $5 per month of AI credits. Startup is $99 per month for small teams and agencies: 4 computers, 32 GB RAM, a 160 GB storage pool, 5 team members, Linux, priority support and $10 per month of AI credits. Scale is $399 per month for production teams: 16 computers, 128 GB RAM, a 640 GB storage pool, 25 team members, Linux plus Windows, custom templates, advanced support and $50 per month of AI credits. Dedicated capacity and embedded runtime volume are arranged by contacting the vendor.

Open-source LLM router with self-healing retries that also routes to Claude/ChatGPT subscriptions

Pricing: Free is $0/month: Manifest Cloud or self-hosted with Docker under the MIT licence, unlimited agents, 10,000 routed requests per month on Cloud and unlimited self-hosted, all providers with no restrictions, subscription providers (Claude, ChatGPT, Gemini), 7-day dashboard retention, Autofix, budget alerts and notifications, and community support via Discord. Pro is $19/month on Manifest Cloud and adds unlimited routed requests, 365-day dashboard retention and basic support. Enterprise is quoted by sales and covers on-premise, VPC or private-cloud deployment, multiple seats and team management, SSO/SAML, audit logs, custom retention and a contract.