Model Not Exist

deepseek-chat and deepseek-reasoner stopped on 2026-07-24. A call that still sends one of those ids fails. The JSON below is a real DeepSeek body from a public log. It was not logged for those two ids after the retirement.

• deepseek-chat
• deepseek-reasoner
• deepseek-flash
• deepseek-v4-pro
• deepseek-v4-flash
• Silent failures after a model shutdown
• DeepSeek error codes
• DeepSeek thinking mode

Error text

{"error":{"message":"Model Not Exist","type":"invalid_request_error","param":null,"code":"invalid_request_error"}}

Quoted from GitHub issue NoFxAiOS/nofx#912, checked October 25, 2026.

The DeepSeek API changelog for 2026-04-24 says the legacy names deepseek-chat and deepseek-reasoner will be discontinued on 2026-07-24. During that period they point to the non-thinking mode and the thinking mode of deepseek-v4-flash, respectively. The V4 preview note says those two names are fully retired and inaccessible after Jul 24th, 2026, 15:59 (UTC Time). The 2026-09-10 changelog says deepseek-v4-flash and deepseek-v4-flash-vision-exp have been retired, and that those names are temporarily routed to V4.1 Flash. It also says to use deepseek-flash. The pricing page lists deepseek-flash (DeepSeek-V4.1-Flash-1017) and deepseek-v4-pro (DeepSeek-V4-Pro-0813). It says the legacy names deepseek-v4-flash and deepseek-v4-flash-vision-exp are still accepted, the models have been retired, and requests are served by DeepSeek-V4.1-Flash-1017. The error-codes page lists 400, 401, 402, 422, 429, 500, and 503. It does not print model_not_found, Model Not Exist, or a JSON body. The WorkBuddy page says `Model Not Found` or `404`: check whether the model id is exactly deepseek-flash. That line is client troubleshooting, not an API body. GitHub issue NoFxAiOS/nofx#912, opened 2025-11-11, logs HTTP 400 from https://api.deepseek.com/v1/chat/completions for model name DeepSeek-V3.2-Exp, with the JSON in the Error text block. GitHub issue AstrBotDevs/AstrBot#7286, opened 2026-04-02, logs the same message, type, and code as a Python error for the prefixed id deepseek/deepseek-reasoner sent to https://api.deepseek.com. That was a prefix bug, and it was before 2026-07-24. GitHub issue aaif-goose/goose#13114 says selecting deepseek-reasoner fails with model_not_found and does not print a JSON body. This page did not find a public log that pastes the JSON above for deepseek-chat or deepseek-reasoner after 2026-07-24. Source: DeepSeek API changelog, checked October 25, 2026.

Affected models

Fix

from openai import OpenAI
client = OpenAI(base_url="https://api.deepseek.com")
response = client.chat.completions.create(
    model="deepseek-flash",
    messages=[{"role": "user", "content": "Hello"}],
    extra_body={"thinking": {"type": "disabled"}},
)

from openai import OpenAI
client = OpenAI(base_url="https://api.deepseek.com")
response = client.chat.completions.create(
    model="deepseek-flash",
    messages=[{"role": "user", "content": "Hello"}],
    extra_body={"thinking": {"type": "enabled"}},
)

import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.deepseek.com" });
const response = await client.chat.completions.create({
  model: "deepseek-flash",
  messages: [{ role: "user", content: "Hello" }],
});

Switch guides

• deepseek-chat to deepseek-v4-flash
• deepseek-reasoner to deepseek-v4-flash

All request errors