Fireworks DeepSeek-V4-Pro (Global)

DeepSeek-V4-Pro is a flagship open-source Mixture-of-Experts model designed for frontier reasoning, advanced coding, and long-context intelligence at scale (up to 1M tokens). It introduces a hybrid attention architecture that dramatically improves long-context efficiency while reducing KV and compute overhead, along with stability and training enhancements for deep multi-step reasoning. It represents a top-tier open-source system for complex agentic workflows, high-precision reasoning, and demanding production workloads.

Provider: All DeepSeek models | Fireworks AI

API Endpoint

https://api.fireworks.ai/inference/v1/chat/completions

Quick Start (Python)

Install: pip install openai

from openai import OpenAI

client = OpenAI(
    api_key="your-fireworks-api-key",
    base_url="https://api.fireworks.ai/inference/v1",
)

response = client.chat.completions.create(
    model="accounts/fireworks/models/deepseek-v4-pro",
    messages=[
        {"role": "user", "content": "Hello, how are you?"},
    ],
    max_tokens=1024,
    temperature=0.7,
)

print(response.choices[0].message.content)

Additional examples: Basic invoke, Streaming

Supported Parameters

ParameterTypeDescription
max_tokensintegerMaximum tokens to generate. (≥1)
temperaturefloatControls randomness. (0–2) Default: 0.7.
top_pfloatNucleus sampling threshold. (0–1) Default: 1.
streambooleanStream response chunks as they are generated. Default: false.
stopstringStop sequence or array of stop sequences.

Feature Guides

Serverless Inference

Pay per token for public open models without managing GPU deployments.

Documentation

OpenAI Compatibility

Use OpenAI-compatible client libraries by changing the API base URL.

Documentation