We use cookies to enhance your experience, provide personalised content, and analyse our traffic; see our Cookie Policy.
deepseek/deepseek-v4-flashA million tokens of context, built for volume
DeepSeek V4 Flash is the fast member of the V4 family, released by DeepSeek on 24 April 2026 alongside V4 Pro. Same one-million-token window, same output of up to 384K, and a thinking mode you switch on per request.
It is made for volume: summarising archives, answering questions over a long text, reading and explaining code, running the steps of an agent. According to DeepSeek its reasoning approaches V4 Pro's. On Tchavi it takes text as input.
On Tchavi you pay with mobile money, no bank card and no KYC. One account and one API key open 80+ models: swap the base URL, keep your OpenAI SDK, and your credits never expire.
A codebase or a stack of long documents in a single call
Long answers without splitting the work
Thinking on or off, request by request
Function calling and structured output for your agents
Write a 60-word WhatsApp announcement for my sandwich restaurant in Cotonou, with today's menu and a catchy opening line.
Here is a 200-page contract: summarise it in 10 key points and cite the source clause for each.
Here is my payment API's code: explain a request's path from entry to the database, and point out the risks.
Start with Flash: same million-token context, same 384K output, and according to DeepSeek reasoning close to Pro. Move to DeepSeek V4 Pro (model: "deepseek-v4-pro") for agentic coding and hard problems where Flash stalls. Test both on your own task.
DeepSeek Chat is the previous generation: 128K of context and 8K of output. V4 Flash gives 1M of context and up to 384K of output, and DeepSeek has announced the retirement of its older identifiers. For a new project, start from Flash.
Up to 1,000,000 tokens in and 384,000 out. Set max_tokens to what you actually need: a long answer burns a lot of credits.
Not on Tchavi: it takes text in and returns text. To attach images, try Qwen3.7 Flash (model: "qwen3.7-flash"), which accepts them.
Swap the base URL for Tchavi's, use your Tchavi key and set model: "deepseek-v4-flash" on /v1/chat/completions. DeepSeek-specific parameters, thinking mode included, are passed through as-is: see DeepSeek's documentation.
Yes. Top up your Tchavi account with Orange Money, MTN MoMo, M-Pesa or another mobile money operator. No bank card is needed. Credits land as soon as the payment is confirmed and are deducted as you use them.
Yes. Keep your current OpenAI SDK, just change the base URL to Tchavi's (shown in the API tab) and use your Tchavi key. The rest of your code stays the same.
import Tchavi from '@tchavi/sdk';
const client = new Tchavi({ apiKey: 'API_KEY' });
const response = await client.chat.completions.create({
model: 'deepseek-v4-flash',
messages: [
{
"role": "user",
"content": "Hello!"
}
],
});
console.log(response.choices[0].message.content);
console.log('Credits used:', response.tchavi.credits_used);Every parameter this model accepts on the API. Check the SDK for full typings.
modelRequiredID of the chat model (e.g. `deepseek-v4-flash`).
messagesRequiredConversation messages. Each item has a `role` (system/user/assistant) and a `content` that's either a plain string or an array of content blocks (`text`, `image_url`, `document_url`).
temperature1Sampling temperature between 0 and 2. Higher values = more random.
max_tokensMaximum number of tokens to generate in the response.
max_completion_tokensOpenAI alias for `max_tokens`. Either field works; if both are sent, `max_completion_tokens` wins.
streamfalseIf true, partial message deltas are streamed as Server-Sent Events.
toolsFunction-calling toolbox. Pass an array of `{ type: "function", function: { name, description, parameters } }`. Returned tool calls land in `choices[].message.tool_calls`. Only models with `supportsTools: true` accept this.
tool_choiceHow the model picks among tools: `"auto"` (default), `"none"` to force a plain text reply, or `{ type: "function", function: { name } }` to force a specific tool. Only with `supportsTools: true`.
response_formatConstrain the output. Pass `{ type: "json_object" }` to force valid JSON (the model still needs a JSON instruction in your prompt). Only models with `supportsJsonMode: true` honour this.
top_p1Nucleus sampling: only tokens in the top `top_p` probability mass are considered.
seedInteger seed for deterministic output. The same seed + identical request yields the same completion when supported by the provider. OpenAI honours it; Anthropic ignores.
n1Number of completions to generate in one call. Defaults to 1; each extra completion is billed.
stream_optionsStreaming options, e.g. `{ "include_usage": true }` to receive a final SSE chunk with usage counts. Effective only when `stream: true`.
stopUp to 4 sequences where the API will stop generating further tokens.
frequency_penalty0Between -2.0 and 2.0. Positive values penalize tokens that have appeared frequently.
presence_penalty0Between -2.0 and 2.0. Positive values encourage the model to talk about new topics.
userEnd-user identifier passed to the upstream provider for abuse / moderation tracking. Should be stable per end-user.
logit_biasPer-token logit bias `{ "<tokenId>": <-100..100>, ... }`. OpenAI-specific; Anthropic ignores silently.
Estimated over 30 days, for a typical message: ~500 input + ~150 output tokens, billed to the thousandth of a credit.