Source: https://aiprimetech.io/models/gpt-6-1-sol/

# GPT-6.1 Sol API

A practical starting point for coding and agent workflows. Connect through AI Prime Tech with the exact model ID `gpt-6.1-sol` and a key from a group that serves it.

<p class="cx-actions"><a class="btn primary" href="/register">Create an API key</a><a class="btn" href="/docs/guides/codex-cli/">Connect Codex CLI</a></p>

## When to use GPT-6.1 Sol

Start with GPT-6.1 Sol for feature work, refactoring and code review when both quality and cost matter. Give it a bounded change, repository conventions and the checks that define completion. Keep the model ID explicit when comparing it with GPT-6 Sol so a migration does not silently change the baseline.

## Model details

| Item | Value |
| --- | --- |
| Developer | OpenAI |
| API model ID | `gpt-6.1-sol` |
| Published context window | 1,050,000 tokens |
| Published maximum output | 128,000 tokens |
| Recommended starting endpoint here | `/v1/responses` |
| Gateway key group for prepaid access | Codex |

The context and output figures are OpenAI's published model specifications, checked on 11 October 2026. Effective limits and supported features can depend on your gateway group and client. Confirm them before sending a large request.

## Pricing and access

OpenAI's published standard text-token rates are **$2.00 per million input tokens** and **$10.00 per million output tokens**, with cached input at **$0.10 per million**. Long-context requests, service tiers and tool charges can change the direct OpenAI price. [Official model reference](https://developers.openai.com/api/docs/models/gpt-6.1-sol).

Those figures describe direct OpenAI pricing. For AI Prime Tech, use the live rate and access terms in your dashboard. Select prepaid **Codex** access and check [purchases](/register?next=/purchase) for any separately offered plans. Do not assume that a Claude plan includes this model.

## Python Responses API example

Set `AIPRIMETECH_API_KEY` in your environment and install the OpenAI Python SDK, then run a small streaming request:

```python
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://aiprimetech.io/v1",
    api_key=os.environ["AIPRIMETECH_API_KEY"],
)
response = client.responses.create(
    model="gpt-6.1-sol",
    input="Reply with one sentence confirming the connection.",
    stream=True,
)
for event in response:
    if event.type == "response.output_text.delta":
        print(event.delta, end="", flush=True)
```

For Codex CLI, set `model = "gpt-6.1-sol"` in the provider configuration from the [setup guide](/docs/guides/codex-cli/). Use the full model ID rather than an assumed family alias.

## Verify the result before scaling

Check your key's model list, run a short prompt and review the request in the dashboard. The model has recent completed Responses requests in the prepaid Codex group, but account access and routing can change. Validate tools or structured output separately if your application needs them.

Compare [GPT-6 model choices](/blog/gpt-6-models-for-codex/), browse [OpenAI-compatible API access](/openai-api/) or return to [Codex API access](/codex-api/).

AI Prime Tech is an independent gateway. OpenAI owns the model name and develops the model; no affiliation or endorsement is implied.