deepseek-v3 API

DeepSeek-V3 is the general-purpose chat model that put DeepSeek on the map in late 2024. It uses a Mixture-of-Experts architecture (671B total parameters, ~37B activated per token, per DeepSeek's own technical report), which is why it delivers quality close to much pricier frontier models while staying fast and inexpensive to serve. It is particularly strong on code generation, mathematics, and Chinese–English bilingual tasks, and its open-weights release made it one of the most independently benchmarked models of its generation. Unlike reasoning-first models, deepseek-v3 answers directly without spending hidden "thinking" tokens, so latency is predictable and the first token arrives quickly — a good fit for interactive chat and high-throughput batch jobs where you don't want to pay for deliberation you didn't ask for. On CaMeL Hub it is billed purely per token: $1.50 per 1M input and $6.00 per 1M output. Since output costs 4x input, prompt-heavy workloads such as summarization and classification are especially economical. DeepSeek has since shipped the V4 series (deepseek-v4-flash and deepseek-v4-pro are also available on CaMeL Hub), but deepseek-v3 remains a stable target if your prompts, evals, or fine-tuned pipelines were built against its behavior and you don't want output drift from a model swap. On CaMeL Hub you can reach it through both the OpenAI-compatible endpoint (/v1/chat/completions) and the Anthropic-compatible endpoint (/v1/messages), so existing tooling for either ecosystem works without a rewrite.

curl https://api.camel-hub.com/v1/chat/completions \
  -H "Authorization: Bearer $CAMEL_HUB_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from openai import OpenAI

client = OpenAI(
    api_key="$CAMEL_HUB_KEY",
    base_url="https://api.camel-hub.com/v1",  # <-- the only change vs. the OpenAI default
)

resp = client.chat.completions.create(
    model="deepseek-v3",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.CAMEL_HUB_KEY,
  baseURL: "https://api.camel-hub.com/v1", // <-- the only change vs. the OpenAI default
});

const resp = await client.chat.completions.create({
  model: "deepseek-v3",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);