GPT 5.6 Sol

gpt-5.6-sol

GPT-5.6 Sol is the flagship model in the OpenAI GPT-5.6 series. It is designed for complex reasoning, coding, and agentic workflows, and is especially powerful in command-line and multi-step coding tasks as well as long-horizon problem solving.

Context window100K

ProviderOpenAI

Released2026/07/10

Playground

Pricing

The same model is available through multiple service channels — choose based on latency, reliability and cost.

Prices in $ / 1M tokens

To pick a channel, add a provider field to the request body, for example "provider": { "channel": "direct" }. Valid values are direct / stable / economical; omit it to use the default channel.

Direct

Direct upstream connection — best when you need native behavior and the full context window.

Input context	Input	Output	Cache read	Cache write
≤ 272K	5.00/M	30.00/M	0.50/M	6.25/M
> 272K	10.00/M	45.00/M	1.00/M	12.50/M

5% off

Preferred

A balanced route optimized for availability and speed — a good default for production traffic.

Input context	Input	Output	Cache read	Cache write
≤ 272K	5.004.75/M	30.0028.50/M	0.500.47/M	6.255.94/M
> 272K	10.009.50/M	45.0042.75/M	1.000.95/M	12.5011.88/M

Overview

Input

Text Image

Output

Text

GPT-5.6 Sol API: Flagship Tier for Cutting-Edge Coding and Complex Agents

GPT-5.6 is a three-tier model series officially released by OpenAI on July 9, 2026. GPT-5.6 Sol is the flagship tier, built for cutting-edge coding, long-horizon agents, design, cybersecurity, and expert knowledge work. Choose Sol when the value of task success justifies flagship capability; hand balanced production work to Terra, and large-scale, cost-conscious traffic to Luna.

On the Artificial Analysis Coding Agent Index, Sol leads the leaderboard with a score of 80; it is also in the top tier on Terminal-Bench 2.1 and BrowseComp. It likewise performs strongly in evaluations such as Agents' Last Exam, which measures long-horizon professional workflows.

SeaWhale AI offers GPT-5.6 Sol through an OpenAI-compatible interface, supporting tool calling, reasoning effort control, streaming output, and multimodal input.

Get API Key · Model ID: gpt-5.6-sol

Why Choose GPT-5.6 Sol

The GPT-5.6 series combines higher per-token work efficiency with clear capability tiering, letting applications match model intelligence to request value instead of paying the same price point for every task. Sol is the most capable tier in this lineup:

Leads the Coding Agent Index — 80 points on the Artificial Analysis Coding Agent Index, ranking first
Persistence and verification capability — suited for agents that need to run for extended periods, self-check, and verify results
Adjustable reasoning effort — allocate compute by task difficulty; simple tasks don't have to pay flagship-level token costs
Multimodal input — supports image and file input, fitting real-world engineering and knowledge work scenarios

Core Capabilities

01 Cutting-Edge Coding and Long-Running Agents

Sol is built for codebase-scale engineering, terminal workflows, complex planning, and high-value tasks that require persistence and verification.

Plan, implement, test, and verify across codebases and terminal environments
Stable tool calling in long-horizon agent loops
Reasoning effort control from low to high

02 Tool Calling and Agent State Management

Send instructions and structured input, enable function calling or web search, and receive standard responses containing usage data. The GPT-5.6 series introduces programmatic tool calling in the Responses API, further reducing round-trip overhead for multi-step tasks.

Function calling and parallel tool execution
Streaming output and incremental tool arguments
Image and file input

03 Stronger Expert Knowledge Work

GPT-5.6 improves end-to-end handling of documents, presentations, spreadsheets, research briefs, and editable artifacts, following source material and visual structure more accurately.

Presentation and editable document generation
Financial and research analysis
Higher fidelity to templates and reference files

04 Cybersecurity, Science, and Research Workflows

For eligible defensive security and research workflows, Sol can be used for security code review, patch validation, scientific analysis, and experiment debugging.

Defensive security analysis and patch validation
Scientific research support
Experiment debugging and system optimization

Best Use Cases

Scenario	Description
Coding agents	Plan, implement, test, and verify changes across codebases and terminal environments
Tool-calling agents	Orchestrate functions, search, files, and intermediate results
Design and front-end work	Generate and refine interfaces, visual specifications, prototypes
Knowledge work	Turn mixed source material into structured documents, tables, and reports
Defensive security	Authorized security code review, threat modeling, vulnerability triage
Tiered production routing	Complex requests go to Sol, daily work to Terra, high-volume traffic to Luna

GPT-5.6 Sol vs. Terra and Luna

The main differences among the three GPT-5.6 tiers lie in the balance of capability, speed, and cost. Start large-scale tasks with Luna, choose Terra for balanced production work, and pick Sol when the value of task success justifies the flagship tier.

Capability	GPT-5.6 Luna	GPT-5.6 Terra	GPT-5.6 Sol
Model ID	`gpt-5.6-luna`	`gpt-5.6-terra`	`gpt-5.6-sol`
Positioning	Fastest, most affordable	Balanced everyday model	Flagship frontier model
Coding Agent Index	74.6	77.4	80
Max output	256K tokens	256K tokens	256K tokens
Input modalities	Text, image	Text, image	Text, image
Best for	Extraction, classification, batch processing	Coding, analysis, copilot	Complex agents, security, science, design

GPT-5.6 was released on July 9, 2026. For specific billing, refer to the real-time price card at the top of the page.

FAQ

Is GPT-5.6 Sol available now? Yes. OpenAI released GPT-5.6 on July 9, 2026, and SeaWhale AI already offers all three tiers: Sol, Terra, and Luna.

What's the difference between Sol, Terra, and Luna? Sol is the flagship tier for the most complex work; Terra balances capability and cost; Luna prioritizes speed, throughput, and economy. All three share the same interface—switching only requires changing the model ID.

On which benchmarks does Sol lead? It ranks first with 80 points on the Artificial Analysis Coding Agent Index and is also in the top tier on Terminal-Bench 2.1 and BrowseComp. On repository-level benchmarks such as SWE-Bench Pro, Claude's Mythos tier still has an advantage, so it's recommended to test against your own workload when choosing.

When should I choose Sol? Choose Sol for complex agents, cybersecurity, science, design, or high-value professional work; choose Terra for general production coding and analysis; start repeatable large-scale tasks with Luna.

Do you support tool calling and streaming? Yes. Function calling, parallel tool execution, streaming output, reasoning effort control, and upstream-supported multimodal input are all available.

What are the context and output limits? Up to 256K output tokens per request. For large outputs, use streaming to avoid request timeouts.

Why Use GPT-5.6 Sol API with SeaWhale AI

One API, three tiers — freely switch between Sol / Terra / Luna within the same account and route by task value
OpenAI-compatible interface — existing SDK code only needs base URL and model ID changes
Unified usage and cost management — track spending and set budgets centrally on one platform
Built for multi-model products — shares the same infrastructure as Claude, Gemini, Qwen, and more

API

API integration

Model IDUse this value as the model in inference requests

gpt-5.6-sol

API KeyBearer token used to authenticate inference requests

Manage API keys

Base URLOpenAI compatible · /chat/completions

https://api.seawhaleai.com/v2/

provider OptionalSelects a service channel; omit it and the system picks the default

"provider": { "channel": "direct" }

gpt-5.6-sol usage examples

SeaWhale AI is compatible with the OpenAI API protocol, so you can call it with the OpenAI SDK or plain HTTP requests. Streaming is enabled by default.

About the provider parameter (optional, a SeaWhale AI extension): most models are served over several channels that differ slightly in price and reliability. Add a provider field to the request body to pick one; omit it and the system selects the default channel — normal calls are unaffected.

provider is not part of the official OpenAI protocol — it is a SeaWhale AI extension that only takes effect on this platform. The OpenAI SDK allows custom fields like this to pass through; see the examples below.

Value	Channel	Best for
`direct`	Direct	The official upstream link, for native behavior and the full context window
`stable`	Preferred	Balanced availability and speed — a good default for production traffic
`economical`	Economy	Cost first, well suited to batch processing and price-sensitive workloads

Available channels and their prices are listed under "Pricing" above (channels vary by model). Additional notes:

Syntax: "provider": { "channel": "direct" }.
If the requested channel is not enabled for that model, the request automatically falls back to the default channel and returns normally — it does not error.
With the official SDKs: in Python pass it via extra_body; in Node.js put it directly on the request object and it passes through. In TypeScript projects, add a // @ts-expect-error line to skip the type check.

cURLPythonNode.js

curl https://api.seawhaleai.com/v2/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer <API_KEY>" \
  -d '{
    "model": "gpt-5.6-sol",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    "provider": { "channel": "direct" },
    "stream": true
  }'
# provider is optional — remove this line to use the default channel

from openai import OpenAI

client = OpenAI(
    base_url="https://api.seawhaleai.com/v2",
    api_key="<API_KEY>",
)

stream = client.chat.completions.create(
    model="gpt-5.6-sol",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"},
    ],
    stream=True,
    # Optional: pick a service channel; omit to use the default
    extra_body={"provider": {"channel": "direct"}},
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

import OpenAI from 'openai'

const client = new OpenAI({
  baseURL: 'https://api.seawhaleai.com/v2',
  apiKey: '<API_KEY>',
})

const stream = await client.chat.completions.create({
  model: 'gpt-5.6-sol',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello!' },
  ],
  stream: true,
  // Optional: pick a service channel; omit to use the default
  // @ts-expect-error provider is a SeaWhale AI extension, not in the OpenAI SDK types
  provider: { channel: 'direct' },
})

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '')
}