DawCode lists 2 models: GPT-4o and Claude 3.5 Sonnet. That’s it. No Gemini, no DeepSeek, no open-source Llama variants. If you need breadth, go elsewhere. If you need those two specific models with Chinese payment methods, DawCode is a viable option.

Pricing Model: ¥4 Free Trial

DawCode uses a free trial model. You get ¥4 credit upon signup. No promo code exists. The “Price/month: $0” and “free_trial: True” fields confirm this is a try-before-you-buy setup. Payment methods are Alipay and WeChat Pay. No credit card or crypto options. Bottom line: ¥4 free credit is enough for about 10-20 API calls depending on output length. It’s a test budget, not a production budget.

Model-by-Model Breakdown

GPT-4o

OpenAI’s flagship multimodal model. DawCode routes it through two groups: “MAX” and “cc-stu”. The platform data shows a latency of 1975ms (about 2 seconds) and a speed rating of 3.0/5. That’s below average. The “MAX” group likely has higher stability. The “cc-stu” group recently saw price increases, per the cons list. If you’re testing GPT-4o for vision tasks or complex reasoning, the ¥4 credit covers roughly 15-20 completions at moderate token counts. Key takeaways:

  • Two route groups available; “MAX” likely more stable
  • 1975ms latency is noticeable for real-time chat
  • ¥4 credit covers limited testing, not sustained use

Claude 3.5 Sonnet

Anthropic’s balanced model. Same routing through MAX and cc-stu groups. The context window is 100,000 tokens — that’s the same across both models per the “Max tokens: 100,000” field. Claude 3.5 Sonnet excels at code generation and structured output. With 100K context, you can feed entire codebases or long documents. The 1975ms latency applies here too. Don’t expect snappy responses. Key takeaways:

  • 100K context window is generous for both models
  • Same latency applies across all routes
  • Claude 3.5 Sonnet is strong for code tasks

Pros & Cons

Pros:

  • ¥4 free credit for testing — no upfront payment
  • Supports Alipay and WeChat Pay
  • MAX and cc-stu routing groups available per model
  • Daily check-in rewards for active users
  • 100% uptime per platform data Cons:
  • Only 2 models — no Gemini, DeepSeek, or open-source options
  • Response speed below average (3.0/5, 1975ms latency)
  • cc-stu group prices recently increased
  • Some packages have low daily quota relative to cost
  • No refund policy or minimum recharge specified

Verdict

DawCode is a niche service. It’s not a platform for broad AI access. It’s a testing ground for GPT-4o and Claude 3.5 Sonnet if you’re in China and need Alipay/WeChat Pay. The 100% uptime is impressive, but the 1975ms latency and 3.0/5 speed rating mean you’ll feel the delay. The ¥4 free credit is genuinely useful for verifying that your code works with these models before committing to a paid plan elsewhere. If you need model diversity, skip DawCode. If you need exactly GPT-4o and Claude 3.5 Sonnet with Chinese payment methods, it works.

FAQ

Can I use DawCode for production workloads?

The ¥4 free credit and lack of specified minimum recharge suggest this is a testing platform. The 1975ms latency and 3.0/5 speed rating make it suboptimal for production real-time use. For batch processing or testing, it’s acceptable.

Which models are rate-limited or unlimited?

The platform data doesn’t specify rate limits per model. Both GPT-4o and Claude 3.5 Sonnet are available through MAX and cc-stu groups. The cc-stu group recently raised prices, which may indicate higher demand or cost pressures.

What happens when my ¥4 free credit runs out?

You’ll need to recharge via Alipay or WeChat Pay. No minimum recharge amount is specified. The platform doesn’t mention automatic top-ups or billing cycles.

Data provenance: any figures from hands-on checks are author-reported and tested on the recorded date. Independent verification is unavailable unless a source is linked.