DeepSeek V4-Flash-0731 inference overview

Price per 1M tokens

$0.13 (input)
$0.07 (cache hit)
$0.28 (output)

Parameters

13B (active)
284B (total)

Context Window

262K

Release Date

Jul 2026

DeepSeek V4-Flash-0731 inference details

DeepSeek V4-Flash-0731 is an updated MoE model from the original DeepSeek V4-Flash release. Its smaller size in the DeepSeek V4 family makes it ideal for fast and efficient coding, reasoning, and agentic workloads.

Created by: 

DeepSeek

License: 

mit

Model card: 

				
					import openai
import weave

# Weave autopatches OpenAI to log LLM calls to W&B
weave.init("<team>/<project>")

client = openai.OpenAI(
    # The custom base URL points to W&B Inference
    base_url='https://api.inference.wandb.ai/v1',

    # Get your API key from https://wandb.ai/authorize
    # Consider setting it in the environment as OPENAI_API_KEY instead for safety
    api_key="<your-apikey>",

    # Optional: Team and project for usage tracking
    project="<team>/<project>",
)

response = client.chat.completions.create(
    model="deepseek-ai/DeepSeek-V4-Flash-0731",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Tell me a joke."}
    ],
)

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

DeepSeek V4-Flash-0731 resources

Screenshot 2025-07-30 at 1.38.51 PM
Guide
DeepSeek-R1 vs OpenAI o1 comparison
Screenshot 2025-07-30 at 1.03.14 PM
Course
AI engineering course: Agents
Inference_logo
Guide
W&B Inference powered by CoreWeave