免费模型
DeepSeek V4 Flash256K context
FREE API ACCESS
The free API key
One login. One API key. Access DeepSeek, MiMo, and GPT model variants for prototypes, scripts, agents, dashboards, and small products.
MODELS
Available now
Seamlessly switch between available models. Call endpoints with low-latency and stable uptime.
免费模型
MiMo V2.5256K context
收费模型
GPT 5.6 Sol272K context · Low / Medium / High / xHigh
收费模型
GPT 5.6 Terra272K context · Low / Medium / High / xHigh
收费模型
GPT 5.6 Luna272K context · Low / Medium / High / xHigh
免费模型
Fusion MoA256K context
收费模型
GPT 5.4272K context · Low / Medium / High / xHigh
收费模型
GPT 5.4 Mini272K context · Low / Medium / High / xHigh
FUSION MOA
Fusion MoA — Mixture of Agents
Fusion MoA orchestrates multiple large language models in parallel, routing each query to the most capable model and synthesizing their outputs into a single, high-quality response. It delivers improved accuracy and broader knowledge coverage — at no extra cost.
QUOTA
Free quota and permanent quota
freeapis tracks real usage, applies model prices, then deducts balances or tokens from the matching free or permanent quota bucket.
After authorizing Google or GitHub, free-first models use the free pool first: 5M tokens per 5 hours and 35M tokens per week.
Redeem codes add permanent quota. GPT models always use it; free-first models use it only after free quota is unavailable or exhausted.
Permanent quota is deducted in real-time USD per million tokens based on model prices.
- GPT 5.6 Sol / GPT 5.5: Input $5.00 / Output $30.00 (per 1M)
- GPT 5.6 Terra / GPT 5.4: Input $2.50 / Output $15.00 (per 1M)
- GPT 5.6 Luna: Input $1.00 / Output $6.00 (per 1M)
- GPT 5.4 Mini: Input $0.75 / Output $4.50 (per 1M)
- Deductions are calculated precisely down to single tokens.
SUPPORTED TOOLS
Codex, Copilot, and Claude Code first
freeapis currently supports Codex, Copilot CLI, and Claude Code integrations. Other coding tools and agent clients are being added over time.
EXAMPLE
Built with freeapis
Open source projects using freeapis to power their AI features.
Chat ApplicationAI-powered chat interface
A modern chat application using DeepSeek V4 Flash for intelligent conversations with real-time responses.
Code AssistantDeveloper productivity tool
An intelligent code assistant powered by the currently available coding models for code generation and debugging support.
FAQ
01 What is freeapis?
A small collection of useful APIs that can be used with one free API key.
02 How do I get a key?
Login with Google or GitHub, open API key, and create your key.
03 Which models are free?
Our key currently provides DeepSeek V4 Flash, MiMo V2.5, and Fusion MoA as free models, while GPT model variants consume paid/permanent quota.
04 How can it stay free?
Free quota is funded through the Reward center, where users can watch ads or complete tasks to earn bonus tokens. Rate limits keep endpoints healthy for everyone.
05 Do I need separate keys?
No. One key is enough for all freeapis endpoints.
06 What are the usage limits?
Free quota is 5,000,000 tokens per 5 hours and 35,000,000 tokens per week. Deductions are calculated precisely based on model prices.
Build with one free key.
Get up and running with DeepSeek, MiMo, and GPT model variants in seconds.
Login and create key