Latest models available from the biggest names in the AI race. Ranked by free credits, daily quotas, and uptime.
Highest verified starter allocations with dependable upstreams and frontier model access.
Highest sustained free quota available. Dual-protocol architecture drops straight into Cursor, Claude Code, and autonomous agent loops.
High balance allowance tailored for multi-model reasoning benchmarks and token-intensive test suites.
Excellent sustained throughput for automated test suites and long-running agent loops.
Specializes in coding agents, developer tooling, and custom endpoints with broad model coverage.
Extensive catalog of frontier models. Verify token throughput with a diagnostic check before you commit.
Unified CLI, editor runtime, and background agent harness with built-in model routing.
Generous grants plus 24-hour credit refreshes — built for continuous heavyweight workflows.
Resilient aggregator gateway designed for multi-model fallback and high concurrency.
Direct drop-in compatibility with Claude Code, Cursor, and OpenAI toolkits.
High-quality router architecture optimized for low-latency tool calls and agent integrations.
Credit allowance plus claimable community pool bursts. Requires an authenticated GitHub account 14+ days old.
Modern developer platform with simple token generation and testing tools.
Targeted for AI developer tools, sandboxing, and rapid prototyping.
Massive token pools and focused relay endpoints to widen your fallback pool.
High-volume token grant ideal for large context processing, summarization, and batch workflows.
Programmatic API relay managed and issued via Telegram bot automation.
Curated access to frontier state-of-the-art models for performance verification.
Multi-provider aggregator with active upstream health verification.
Lightweight relay suited to endpoint testing and small applications.
Straightforward free model experimentation endpoint.
Zero-signup relays and community pools. Perfect for quick evaluation before committing.
Zero-signup gateway from OpenCode. Best paired with header User-Agent: opencode/1.18.18.
public • deepseek-v4-flash, mimo-v2.5, glm-5.1Completely free model access for testing and evaluation.
Resilient reverse gateway fallback with a generous default free quota.
Fast Chinese mainland routing for regional inference workloads.
Coding assistant agent harnesses with developer invite grants.
Low-latency inference testing and evaluation tier.
Regional Vietnamese routing for free trial model access.
Campaign reward balance increments on top of the starting grant.
Low-latency Chinese mainland failover endpoint.
Credit allocation weighted toward image and diffusion models.
Aggregator endpoint routing across multiple underlying model providers.
Low-latency regional routing with a free quota tier.
Lightweight failover endpoint for backup routing.
Subscription-based aggregator offering referral reward balance increments.
Scan models, check response speeds, test prompts, and export configurations — all in one place. No credentials stored.
Open AI Checker