OpenAIgpt-6.2-sol
OpenAI's go-to model for everyday work, from coding to running agents at scale
$0.5/ 1M input tokens-75%$2official price
- 1.05M context
- Up to 128K output
- Effort low to max
- Reads images
- Modality
- Text
- Endpoint
- /v1/responses
- Model ID
- gpt-6.2-sol
- Provider
- OpenAI
GPT-6.2 Sol API overview
Documented endpoint
Requests go to /v1/responses, or to /v1/chat/completions in the messages format. /v1/messages is not open for OpenAI models.
Model provider
GPT-6.2 Sol is published by OpenAI and is called through the API model ID gpt-6.2-sol.
Output and limits
1,050,000-token context, up to 922,000 of it prompt; up to 128,000 output tokens; text and images in, text out.
GPT-6.2 Sol API pricing
GPT-6.2 Sol is billed 75% below the official list price on every metered dimension below.
| Item | ArgoLink rate | Official price | You save |
|---|---|---|---|
| Input tokens | $0.5 / 1M tokens | -75% | |
| Cached input tokens | $0.025 / 1M tokens | -75% | |
| Cache write tokens | $0.625 / 1M tokens | -75% | |
| Output tokens | $2.5 / 1M tokens | -75% |
Reasoning tokens are billed as output tokens. Fast mode (service_tier priority) bills every token at 2× these rates. When a request's prompt is over 272K tokens, cached tokens included, the whole request bills its output at 1.5× these rates and every other token at 2×. Fast mode and the long-context rates multiply.
Compared with other platforms
| Platform | Input tokens / 1M tokens | Output tokens / 1M tokens |
|---|---|---|
| ArgoLink | $0.5 | $2.5 |
| Official | $2 -75% | $10 -75% |
| OpenRouter | $2 -75% | $10 -75% |
Each platform's list price per 1M tokens, as its own page showed it. The green figure is how much less ArgoLink charges. Checked on
GPT-6.2 Sol API reference
Quick start
Create an API key in the console, then use the Base URL, endpoint, and request body shown on this page.
curl -sS https://argolink.io/v1/responses \
-H "Authorization: Bearer $ARGOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6.2-sol",
"input": "Explain what this model is good at in one sentence.",
"stream": false
}'
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["ARGOLINK_API_KEY"],
base_url="https://argolink.io/v1",
)
response = client.responses.create(
model="gpt-6.2-sol",
input="Explain what this model is good at in one sentence.",
)
print(response.output_text)
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.ARGOLINK_API_KEY,
baseURL: "https://argolink.io/v1",
});
const response = await client.responses.create({
model: "gpt-6.2-sol",
input: "Explain what this model is good at in one sentence.",
});
console.log(response.output_text);
Protocols
Responses
Recommended: the model's own format
The model's own format, and the one Codex speaks: send input as a string or as messages, and read the answer from the output_text items of output.
curl https://argolink.io/v1/responses \
-H "Authorization: Bearer $ARGOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6.2-sol",
"input": "Compare optimistic and pessimistic locking in three bullet points.",
"reasoning": {"effort": "high"},
"max_output_tokens": 8000
}'
curl https://argolink.io/v1/responses \
-H "Authorization: Bearer $ARGOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6.2-sol",
"input": [{
"role": "user",
"content": [
{"type": "input_text", "text": "What does this chart show?"},
{"type": "input_image", "image_url": "https://example.com/chart.png"}
]
}]
}'
| Field | Value |
|---|---|
input | a string, or messages whose content mixes input_text parts and input_image parts, each a public HTTPS URL or a data URL |
instructions | the system prompt |
reasoning.effort | low · medium · high · xhigh · max |
max_output_tokens | up to 128,000 |
stream | true streams server-sent events |
service_tier | priority (or fast) asks for Fast mode, billed at 2× the standard rates |
Reasoning tokens count toward max_output_tokens. Use the Responses API for tool calling; Chat Completions supports this model without tools. A reasoning effort of none is rejected: use low in place of none or minimal.
Chat Completions
For clients built on the messages format
The same model through the messages format that OpenAI-compatible clients send; the fields map one to one.
curl https://argolink.io/v1/chat/completions \
-H "Authorization: Bearer $ARGOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-6.2-sol",
"messages": [{"role": "user", "content": "Compare optimistic and pessimistic locking in three bullet points."}],
"reasoning_effort": "high"
}'
| Field | Value |
|---|---|
messages | system, user, assistant and tool turns; a user turn may carry image_url parts, each a public HTTPS URL or a data URL |
reasoning_effort | the same values as reasoning.effort |
max_completion_tokens | up to 128,000; max_tokens works too |
Codex
Responses, set up once in config.toml
Codex CLI and Codex Desktop read the same file. Save it, then open a new Codex session.
model = "gpt-6.2-sol"
model_provider = "argolink"
[model_providers.argolink]
name = "ArgoLink"
base_url = "https://argolink.io/v1"
wire_api = "responses"
requires_openai_auth = false
experimental_bearer_token = "YOUR_API_KEY"
GPT-6.2 Sol full request parameters
| Parameter | Type | Requirement | What it does |
|---|---|---|---|
model | string | required | The catalog identifier for this page: gpt-6.2-sol. |
input | string or array | required | The prompt as a string, or messages whose content mixes input_text and input_image parts; an image is a public HTTPS URL or a data URL. |
instructions | string | optional | The system prompt for this request. |
reasoning.effort | string | optional | Reasoning depth: low · medium · high · xhigh · max. |
max_output_tokens | integer | optional | Caps the answer, reasoning included: up to 128,000. |
tools · tool_choice | array · string or object | optional | Function tools and how the model picks one. |
stream | boolean | optional | true streams server-sent events such as response.output_text.delta. |
service_tier | string | optional | priority (fast is the same) asks for Fast mode and bills every token at 2× the standard rates; leave it out for the standard tier. |
usage | object | response field | input_tokens with input_tokens_details.cached_tokens, and output_tokens with output_tokens_details.reasoning_tokens. |
GPT-6.2 Sol API compared with similar models
| Model | Model ID | Provider | Rate | Best for |
|---|---|---|---|---|
| GPT-6.2 Sol | gpt-6.2-sol | OpenAI | $0.5 / 1M input tokens | OpenAI's go-to model for everyday work, from coding to running agents at scale |
| GPT-6 Astra | gpt-6-astra | OpenAI | $2.3 / 1M input tokens | OpenAI's most capable model, built for the hardest end-to-end work |
| GPT-5.6 Terra | gpt-5.6-terra | OpenAI | $0.46 / 1M input tokens | The GPT-5.6 model that balances intelligence and cost |
| GPT-5.6 Sol | gpt-5.6-sol | OpenAI | $0.92 / 1M input tokens | The GPT-5.6 flagship for complex professional work |
More models from this provider
- GPT-6 AstraOpenAI's most capable model, built for the hardest end-to-end work$2.3 / 1M input tokens
- GPT-5.6 TerraThe GPT-5.6 model that balances intelligence and cost$0.46 / 1M input tokens
- GPT-5.6 SolThe GPT-5.6 flagship for complex professional work$0.92 / 1M input tokens
- GPT-5.5A flagship for coding and the most complex professional work$1.15 / 1M input tokens
Other models in this modality
- Claude Opus 5For complex agentic coding and enterprise work$1.75 / 1M input tokens
- Claude Fable 5.1Anthropic's model for demanding reasoning and long-horizon agentic work$3.5 / 1M input tokens
- Claude Fable 5Anthropic's Fable model for long-running agentic work$3.5 / 1M input tokens
- Grok 4.6xAI's frontier model for coding, agentic tasks and knowledge work$0.6 / 1M input tokens
GPT-6.2 Sol API FAQ
How do I call GPT-6.2 Sol?
Create an API key in the console and POST https://argolink.io/v1/responses with model set to gpt-6.2-sol and your key in the Authorization header as a Bearer token. /v1/chat/completions works too; Anthropic Messages is not open for OpenAI models.
How much does GPT-6.2 Sol cost?
$0.5 per million input tokens and $2.5 per million output tokens, 75% below the official rates. Cached input costs $0.025 per million. Reasoning tokens are billed as output tokens. Fast mode (service_tier priority) bills every token at 2× these rates. When a request's prompt is over 272K tokens, cached tokens included, the whole request bills its output at 1.5× these rates and every other token at 2×. Fast mode and the long-context rates multiply.
What are GPT-6.2 Sol's context and output limits?
1,050,000-token context, up to 922,000 of it prompt; up to 128,000 output tokens; text and images in, text out.
How do I control GPT-6.2 Sol's reasoning?
Set reasoning.effort to low, medium, high, xhigh or max.
Can GPT-6.2 Sol read images?
Yes: send them as input_image parts, each a public HTTPS URL or a data URL, on /v1/responses.