{"models":[{"name":"deepseek-v4-1-flash","vendor":"DeepSeek","pricing":{"kind":"perToken","inputUsdPerMillion":0.10500765,"outputUsdPerMillion":0.4200306,"cacheReadUsdPerMillion":0.002100153,"cacheWriteUsdPerMillion":0},"endpoints":["openai"],"displayName":"DeepSeek-V4.1-Flash","description":"The smallest model in DeepSeek's new architecture family, with native image understanding. It runs in thinking mode by default and also offers a non-thinking mode; DeepSeek reports it outperforms its earlier DeepSeek-V4-Pro flagship on benchmarks.","contextLength":1000000,"maxOutputTokens":384000,"releasedAt":"2026-09-10","knowledgeCutoff":null,"inputModalities":["text","image"],"outputModalities":["text"],"tokens7d":2900000000,"listPricing":{"kind":"perToken","inputUsdPerMillion":0.3,"outputUsdPerMillion":1.2,"cacheReadUsdPerMillion":0.006,"cacheWriteUsdPerMillion":0},"discountPercent":65,"health":{"windowEnd":"2026-11-15T09:35:11.000Z","successRate":99.8,"latencyMs":9125,"tokensPerSecond":93.4,"hourly":[100,100,100,100,100,99.8,100,100,100,100,100,100,100,100,100,100,100,100,100,95.4,100,100,100,100]},"openWeights":{"license":"MIT","licenseUrl":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash/blob/main/LICENSE","weightsUrl":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash","osiApproved":true},"timeOfDay":{"timeZone":"Asia/Shanghai","peakHours":[{"weekdays":[1,2,3,4,5],"fromHour":9,"toHour":12},{"weekdays":[1,2,3,4,5],"fromHour":14,"toHour":18}],"peak":{"kind":"perToken","inputUsdPerMillion":0.2100153,"outputUsdPerMillion":0.8400612,"cacheReadUsdPerMillion":0.004200306,"cacheWriteUsdPerMillion":0},"offPeakList":{"kind":"perToken","inputUsdPerMillion":0.15,"outputUsdPerMillion":0.6,"cacheReadUsdPerMillion":0.003,"cacheWriteUsdPerMillion":0}},"version":{"releases":[{"name":"DeepSeek-V4.1-Flash-1017","releasedAt":"2026-10-17","notesUrl":"https://api-docs.deepseek.com/news/news261017/"},{"name":"DeepSeek-V4.1-Flash","releasedAt":"2026-09-10","notesUrl":"https://api-docs.deepseek.com/news/news260910/"}],"vendorApiName":"deepseek-flash","responseModel":"deepseek-v4.1-flash-1017","aliases":["deepseek-v4-1-flash-1017","deepseek-v4.1-flash-1017"]}},{"name":"glm-5.3","vendor":"Z.ai","pricing":{"kind":"perToken","inputUsdPerMillion":0.9800714,"outputUsdPerMillion":3.0802244,"cacheReadUsdPerMillion":0.18201326,"cacheWriteUsdPerMillion":0},"endpoints":["openai"],"displayName":"GLM-5.3","description":"Z.ai's flagship GLM model for complex software engineering and agentic work. It keeps GLM-5.2's base model, with post-training gains in coding, long-horizon tasks and cybersecurity, and currently accepts text input only.","contextLength":1000000,"maxOutputTokens":128000,"releasedAt":"2026-08-18","knowledgeCutoff":null,"inputModalities":["text"],"outputModalities":["text"],"tokens7d":null,"listPricing":{"kind":"perToken","inputUsdPerMillion":1.4,"outputUsdPerMillion":4.4,"cacheReadUsdPerMillion":0.26,"cacheWriteUsdPerMillion":0},"discountPercent":30,"health":{"windowEnd":"2026-11-15T09:35:11.000Z","successRate":100,"latencyMs":2079,"tokensPerSecond":22.1,"hourly":[100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100]},"openWeights":{"license":"GLM-5.3 License","licenseUrl":"https://huggingface.co/zai-org/GLM-5.3/blob/main/LICENSE","weightsUrl":"https://huggingface.co/zai-org/GLM-5.3","osiApproved":false},"timeOfDay":null,"version":null},{"name":"glm-5.3-flash","vendor":"Z.ai","pricing":{"kind":"perToken","inputUsdPerMillion":0.10500765,"outputUsdPerMillion":0.3500255,"cacheReadUsdPerMillion":0.02100153,"cacheWriteUsdPerMillion":0},"endpoints":["openai"],"displayName":"GLM-5.3-Flash","description":"The first natively multimodal model in Z.ai's GLM-5 series, taking text, image, video and file input. Z.ai positions it as stronger than GLM-5.2 at a much lower cost.","contextLength":1000000,"maxOutputTokens":128000,"releasedAt":"2026-08-26","knowledgeCutoff":null,"inputModalities":["text","image","video","file"],"outputModalities":["text"],"tokens7d":null,"listPricing":{"kind":"perToken","inputUsdPerMillion":0.15,"outputUsdPerMillion":0.5,"cacheReadUsdPerMillion":0.03,"cacheWriteUsdPerMillion":0},"discountPercent":30,"health":{"windowEnd":"2026-11-15T09:35:11.000Z","successRate":100,"latencyMs":3516,"tokensPerSecond":88.5,"hourly":[100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100]},"openWeights":{"license":"MIT","licenseUrl":"https://huggingface.co/zai-org/GLM-5.3-Flash/blob/main/LICENSE","weightsUrl":"https://huggingface.co/zai-org/GLM-5.3-Flash","osiApproved":true},"timeOfDay":null,"version":null},{"name":"grok-4.7","vendor":"xAI","pricing":{"kind":"perToken","inputUsdPerMillion":0.199588,"outputUsdPerMillion":0.598764,"cacheReadUsdPerMillion":0.049897,"cacheWriteUsdPerMillion":null},"endpoints":["openai"],"displayName":"Grok 4.7","description":"xAI's most capable model, for coding, agentic tasks and knowledge work, released in September 2026 as the successor to Grok 4.6. It takes text and image input with a 500K-token context window and offers four reasoning effort levels, from low to xhigh.","contextLength":500000,"maxOutputTokens":null,"releasedAt":null,"knowledgeCutoff":"2026-05","inputModalities":["text","image"],"outputModalities":["text"],"tokens7d":1500000,"listPricing":{"kind":"perToken","inputUsdPerMillion":2,"outputUsdPerMillion":6,"cacheReadUsdPerMillion":0.5,"cacheWriteUsdPerMillion":null},"discountPercent":90,"health":{"windowEnd":"2026-11-15T09:35:11.000Z","successRate":100,"latencyMs":5882,"tokensPerSecond":47.8,"hourly":[100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100]},"openWeights":null,"timeOfDay":null,"version":null},{"name":"kimi-k3","vendor":"Moonshot","pricing":{"kind":"perToken","inputUsdPerMillion":2.100153,"outputUsdPerMillion":10.500765,"cacheReadUsdPerMillion":0.2100153,"cacheWriteUsdPerMillion":2.100153},"endpoints":["openai"],"displayName":"Kimi K3","description":"Moonshot AI's flagship open-weight model, with 2.8 trillion parameters built on Kimi Delta Attention and Attention Residuals. It has native visual understanding and is designed for long-horizon coding, knowledge work and deep reasoning.","contextLength":1000000,"maxOutputTokens":1048576,"releasedAt":"2026-07-16","knowledgeCutoff":null,"inputModalities":["text","image","video"],"outputModalities":["text"],"tokens7d":null,"listPricing":{"kind":"perToken","inputUsdPerMillion":3,"outputUsdPerMillion":15,"cacheReadUsdPerMillion":0.3,"cacheWriteUsdPerMillion":3},"discountPercent":30,"health":{"windowEnd":"2026-11-15T09:35:11.000Z","successRate":100,"latencyMs":2591,"tokensPerSecond":9.7,"hourly":[100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100,100]},"openWeights":{"license":"Kimi K3 License","licenseUrl":"https://huggingface.co/moonshotai/Kimi-K3/blob/main/LICENSE","weightsUrl":"https://huggingface.co/moonshotai/Kimi-K3","osiApproved":false},"timeOfDay":null,"version":null}],"fetchedAt":"2026-11-15T09:35:11.781Z","topup":{"minUsd":5,"presetsUsd":[5,10,20,50,100]}}