Skip to main content

Deep Agents Code

Use Deep Agents Code (dcode) with Divyam for text-file edits and shell commands.

Before you start

You need a Divyam account and an API key. If you don't have them yet, follow Getting Started first.

Replace the shortened divyam-sw-v1-... shown in examples with your complete API key.

The examples use divyam-as/gpt-6-sol. For another model, select its current ID in your harness or copy it from Models and pricing, and check its context limit and supported features.

1. Install Deep Agents Code and set your key​

Install Deep Agents Code using the official instructions, then export your key in the terminal where you will start dcode:

export DIVYAM_API_KEY="divyam-sw-v1-..."

2. Configure Divyam​

divyam-as/

Model ID: divyam-as/gpt-6-sol. Context window: 1,050,000 tokens. Models and pricing

Profile connection

This setup redirects the built-in openai connection in your current Deep Agents profile to Divyam. Later sessions using that profile and provider use these settings. It does not register a separate provider named Divyam.

Merge this into ~/.deepagents/config.toml, creating the directory and file if needed. If you set DEEPAGENTS_HOME, use config.toml inside that directory instead. See the profile location documentation.

[models.providers.openai]
models = ["divyam-as/gpt-6-sol"]
base_url = "https://api.preview.divyam.ai/v1"
api_key_env = "DIVYAM_API_KEY"

[models.providers.openai.params]
use_responses_api = false

[models.providers.openai.profile]
max_input_tokens = 200000

use_responses_api = false selects Chat Completions. If you saved an OpenAI key in dcode, use /auth to replace it with your Divyam key.

3. Start an interactive session​

From your project folder, run:

dcode --model openai:divyam-as/gpt-6-sol
In Deep Agents Code's composer
Hello, what are your capabilities?
Paste this message, then send it.
Feature limits

Image reads through Deep Agents Code's file tools can return a server error. Use text files.

Check your requests​

Open the Divyam console and select Logs. Your requests appear there within a few seconds, newest first, with their status, token counts and latency. Usage and Credits & Billing catch up within a few minutes.

Local cost estimates

Some clients display zero when model prices are missing. Check Credits & Billing for Divyam charges. Local dollar limits depend on the client's price and usage data.

For filters, cached-token counts and the requested-versus-routed view, see Logs and usage.