We use cookies to enhance your experience, provide personalised content, and analyse our traffic; see our Cookie Policy.
deepseek/deepseek-v4-proHeavy reasoning and coding over a 1M-token context
DeepSeek V4 Pro is the flagship of the V4 family, released by DeepSeek on 24 April 2026: 1.6T parameters with 49B active, according to DeepSeek, over a one-million-token window with up to 384K of output.
It targets hard work: coding agents that run for a long time, refactoring a large codebase, maths and multi-step planning. DeepSeek highlights its agentic capabilities and its reasoning. Text input, with tools, JSON mode and a thinking mode.
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.
An entire codebase readable in a single call
Very long answers with no splitting
Thinking on or off, depending on difficulty
Stronger agentic capabilities, according to DeepSeek
Function calling and structured output
Build me a step-by-step plan to migrate a monolith to microservices, with the order of extractions and the risks at each step.
Review this 800-line file and suggest the three highest-impact refactors, with before and after code.
Act as a research agent: outline the steps to compare three vector databases for our use case, then list the tests to run.
Flash keeps the same million-token context and, according to DeepSeek, reasoning close to Pro, for volume work. Pick Pro (model: "deepseek-v4-pro") for coding agents and hard problems. The safe route: try Flash first, step up if needed.
Kimi K3 (model: "kimi-k3") also reads images and can write up to 943K tokens in one go. V4 Pro is text-only, with 384K of output, aimed at code and agents. If you only send text, try V4 Pro first.
Up to 1,000,000 tokens in and 384,000 out. Cap max_tokens at what you really need.
No: DeepSeek V4 Pro has no vision and Tchavi sends it text only. For images, pick Kimi K3 or Qwen3.8 Max.
Swap the base URL for Tchavi's, use your Tchavi key and set model: "deepseek-v4-pro" on /v1/chat/completions. DeepSeek-specific parameters, thinking mode included, are passed through as-is.
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-pro',
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-pro`).
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.