The short answer
- No free tier: the Claude API has usage tiers with monthly spend caps. There is no tier with free recurring tokens.
- New accounts: new organizations can start in an Evaluation tier with lower limits. This is a limit level, not free usage.
- Starter credit: some third-party pages say that new accounts get a small one-time credit. We could not confirm an amount on an official Anthropic page.
- Free Claude chat: the free plan on claude.ai is a chat product. It does not include API access.
Legitimate ways to get Claude API credits
| Option | What you get | Conditions |
|---|---|---|
| Claude Max plan | $100 of API credit per month | Paid subscription. Claim after 7 days on the plan |
| Claude Ultra plan | $200 of API credit per month | Paid subscription. Claim after 7 days on the plan |
| Claude Team plan | $20 per Standard seat and $100 per Premium seat, pooled, up to $500 per month | Paid subscription. An owner links a Console organization |
| Pay as you go | Credits that you buy in the Claude Console | Billed per token. Tier and spend cap apply |
| Cloud platforms | Claude models through a cloud provider | Billed by the cloud provider. Check its own trial offers |
The plan credits are included in a paid plan. They are not free for a person on the Free or Pro plan. Free, Pro and Enterprise plans are not eligible.
- The credits cover the Claude API, Claude Managed Agents, the Claude Agent SDK and the Console playground.
- They do not cover Claude Code or extra usage in the Claude apps.
- They do not apply on any cloud platform.
- Unused credits expire at the end of each billing cycle. They do not roll over.
Make paid Claude credits last longer
If you need real Claude models, the next best thing to free is to spend less per task. The Claude documentation describes three methods.
- Choose a smaller model for simple tasks. Use a Haiku model for short classification or extraction jobs.
- Use prompt caching for repeated context, such as system prompts, tool definitions and long documents. Cached input tokens cost a fraction of the base input price.
- Use the Message Batches API for work that does not need an immediate answer. Anthropic says that it costs 50 percent less.
- Set a spend limit on a workspace in the Claude Console. Requests stop when the workspace reaches the limit.
A common pattern is to mix both worlds. Send hard tasks to a paid Claude model. Send drafts, tests and bulk jobs to free models through a gateway.
Claude models on free routes
Some third-party gateways give a small shared allowance that can include a Claude model. These allowances come from the gateway, not from Anthropic. They are small, shared with other users and can end at any time. The table shows the Claude rows in today's free catalog.
| Model | Free on | Context | Free limits (best offer) | Tools | Vision |
|---|---|---|---|---|---|
| Claude Haiku 4.5 | ACLIDE | - | €20/month shared internal usage credits · no card | No | No |
Live data from the FreeLLMAPI catalog (version 2026.11.19, checked 19 November 2026). Free limits change often. The catalog is checked every day.
Do not plan a product on these rows. Use them for short tests only. For steady free use, run Claude-format code on other free models, as the next section shows.
A free endpoint in the Claude API format
FreeLLMAPI speaks the Anthropic Messages API on /v1/messages. Code built for Claude, including the official Anthropic SDKs, can send requests to it. The router translates each request and sends it to a free model, for example Gemini, GPT-OSS, Qwen or GLM. You do not get Claude models. You get the Claude API format on free models.
Anthropic SDK pointed at FreeLLMAPI
import anthropic
client = anthropic.Anthropic(
base_url="http://localhost:3001",
api_key="<unified-key>",
)
msg = client.messages.create(
model="claude-sonnet-4-5", # mapped to a free model on the Keys page
max_tokens=256,
messages=[{"role": "user", "content": "Say hello in one line."}],
)
print(msg.content[0].text)- Install FreeLLMAPI and add two or more free provider keys.
- Open the Keys, Anthropic tab.
- Map each Claude family (
default,opus,sonnet,haiku) toautoor to a free model. - Set the base URL to the server origin. Anthropic clients add
/v1/messagesthemselves. - Send your unified key in
x-api-keyor as a Bearer token.
Streaming, system prompts, tool use and image input work through the same router. For Claude Code, read How to use Claude Code for free.
Free models for Claude-format code
Claude-style apps often use tools and long prompts. The table shows the strongest free models with tool calling and 128K or more context, from today's catalog.
| Model | Free on | Context | Free limits (best offer) | Tools | Vision |
|---|---|---|---|---|---|
| GLM 5.3 | UnoRouter | 1M | 1 req/min 1 req/min per model | Yes | No |
| GLM 5.3 Flash | NVIDIA NIM, UnoRouter | 1M | 40 req/min 40 RPM | Yes | Yes |
| Gemini 3.8 Flash | Google AI Studio | 1M | Recurring provider quota | Yes | Yes |
| Qwen3.8 Flash Next | AMD Radeon Cloud | 262K | Recurring daily account allowance | Yes | No |
| Qwen3.8 Flash | ElectronHub | 1M | $0.25/week shared recurring credit | Yes | Yes |
| Gemini 3.7 Flash | Google AI Studio | 1M | 10 req/min 20 req/day 250,000 tokens/min ~3M | Yes | Yes |
| GPT-5.6 Luna | ElectronHub | 1M | $0.25/week shared recurring credit | Yes | Yes |
| Dots3 Note Preview | Kilo Gateway, OpenRouter | 512K | 20 req/min 50 req/day 50 req/day account-wide (1,000 after a $10 purchase) | Yes | Yes |
| DeepSeek V4 Flash | AMD Radeon Cloud, BazaarLink, ElectronHub | 1M | 10 req/min 50 req/day Shared rate limit | Yes | No |
| Qwen3.8 27B | ElectronHub, Groq, OVH AI Endpoints | 262K | 30 req/min 250 req/day 70,000 tokens/min ~6M | Yes | Yes |
Showing 10 of 109 matching free models. See all models.
Live data from the FreeLLMAPI catalog (version 2026.11.19, checked 19 November 2026). Free limits change often. The catalog is checked every day.
Quality is not the same as Claude. Test your prompts on two or three models before you choose one. Free tiers also have daily caps, so the router can move to a smaller model late in the day.
Frequently asked questions
Is the Claude API free?
No. Anthropic has no free tier for the Claude API. Requests use credits that you buy, or monthly credits that come with the paid Max, Ultra and Team plans.
How do I get a free Claude API key?
You can create an API key in the Claude Console at no cost, but requests need credit. There is no legitimate key with free recurring Claude usage.
Does Claude Pro include API credits?
No. Anthropic says Free, Pro and Enterprise plans are not eligible for the monthly API credits. Max, Ultra and Team plans are eligible.
Can I use Claude Code with the plan API credits?
No. The monthly API credits do not cover Claude Code. To run Claude Code on free models, use an Anthropic-compatible gateway such as FreeLLMAPI.
Is there a free API that works like Claude?
Yes. FreeLLMAPI serves the Anthropic Messages format on /v1/messages and routes each request to free models such as Gemini, GPT-OSS or Qwen. The models are not Claude.
Related guides
- How to use Claude Code for free
- Is there a free GPT API?
- Best free coding model APIs
- Free LLM API for AI agents
- Free LLM API with function calling
Sources
- Claude Platform docs: API credits for Max, Ultra and Team plans (checked 20 November 2026)
- Claude Platform docs: Rate limits (checked 20 November 2026)
- Claude Help Center: Monthly API credits for Max and Team plans (checked 20 November 2026)
- FreeLLMAPI docs: Anthropic and Claude clients (checked 20 November 2026)