Skip to content

Commit 69c23bc

Browse files
smakoshsmakosh
committed
feat(models): add OpenAI GPT-5.6 sol/terra/luna
Adds OpenAI's GPT-5.6 family released 2026-07-09 with openai and azure provider mappings following the gpt-5.5 pattern: - gpt-5.6-sol: frontier, $5/$30 per 1M tokens (cached $0.50) - gpt-5.6-terra: balanced, $2.50/$15 per 1M tokens (cached $0.25) - gpt-5.6-luna: cost-optimized, $1/$6 per 1M tokens (cached $0.10) All three: 1.05M context, 128K max output, reasoning, vision, tools, web search, structured outputs, flex/priority tiers (priority 2x per OpenAI's priority pricing table). Also adds a changelog entry with OG image, refreshes the curated model category lists, and swaps the Code landing flagship showcase to gpt-5.6-sol. Claude-Session: https://claude.ai/code/session_01C1G9vcptkguozAPhQS9Xec
1 parent bae7652 commit 69c23bc

5 files changed

Lines changed: 257 additions & 1 deletion

File tree

‎apps/code/src/components/CodingModelsShowcase.tsx‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -25,7 +25,7 @@ const RECOMMENDED_MODELS: RecommendedModel[] = [
2525
{ id: "qwen3-coder", category: "cheap" },
2626
{ id: "deepseek-v4-pro", category: "cheap" },
2727
{ id: "claude-opus-4-8", category: "flagship" },
28-
{ id: "gpt-5.5", category: "flagship" },
28+
{ id: "gpt-5.6-sol", category: "flagship" },
2929
{ id: "gemini-3.1-pro-preview", category: "flagship" },
3030
];
3131

‎apps/ui/public/changelog/gpt-5-6-models.png‎

2.08 MB
Loading

‎apps/ui/src/content/changelog/2026-07-09-gpt-5-6-models.md‎

Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
---
2+
id: "64"
3+
slug: "gpt-5-6-models"
4+
date: "2026-07-09"
5+
title: "GPT-5.6 Sol, Terra & Luna Now Available"
6+
summary: "OpenAI's new GPT-5.6 family is live on LLM Gateway from day one: Sol for frontier reasoning at $5/$30 per 1M tokens, Terra balancing intelligence and cost at $2.50/$15, and Luna for high-volume workloads at $1/$6 — all with a 1.05M-token context window, reasoning, vision, tool calling, and web search."
7+
image:
8+
src: "/changelog/gpt-5-6-models.png"
9+
alt: "GPT-5.6 Sol, Terra, and Luna models now available on LLM Gateway"
10+
width: 1536
11+
height: 1024
12+
---
13+
14+
Picking an OpenAI model used to mean trading capability against cost across scattered version numbers. The new **GPT-5.6** family makes the trade-off explicit — three models, one lineup — and all three are available on LLM Gateway today, the day OpenAI released them.
15+
16+
## Three Models, One Family
17+
18+
| Model | Best for | Input / 1M | Cached input / 1M | Output / 1M |
19+
| --------------- | ---------------------------------------------------- | ---------- | ----------------- | ----------- |
20+
| `gpt-5.6-sol` | Frontier reasoning and complex professional work | $5.00 | $0.50 | $30.00 |
21+
| `gpt-5.6-terra` | Balanced intelligence and cost | $2.50 | $0.25 | $15.00 |
22+
| `gpt-5.6-luna` | High-volume tasks like classification and extraction | $1.00 | $0.10 | $6.00 |
23+
24+
Every model in the family shares the same capabilities:
25+
26+
- **Context window**: 1,050,000 tokens, with 128,000 max output tokens
27+
- **Knowledge cutoff**: February 16, 2026
28+
- Reasoning, vision, tool calling, web search, and structured outputs
29+
- **Flex** (50% cheaper) and **Priority** (2x, low-latency) service tiers
30+
31+
## Getting Started
32+
33+
All three models work through the unified API right now:
34+
35+
```bash
36+
curl -X POST https://api.llmgateway.io/v1/chat/completions \
37+
-H "Authorization: Bearer $LLM_GATEWAY_API_KEY" \
38+
-H "Content-Type: application/json" \
39+
-d '{
40+
"model": "openai/gpt-5.6-sol",
41+
"messages": [{"role": "user", "content": "Hello GPT-5.6!"}]
42+
}'
43+
```
44+
45+
Swap `gpt-5.6-sol` for `gpt-5.6-terra` or `gpt-5.6-luna` to move down the cost curve without touching anything else in your integration. Available on all plans.
46+
47+
---
48+
49+
**[Browse the models →](https://llmgateway.io/models)** | **[Quick start →](https://docs.llmgateway.io)**

‎apps/ui/src/lib/model-category-filters.ts‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -60,6 +60,8 @@ export const curatedCategoryModelIds: Record<
6060
"gpt-5.2-codex",
6161
"gpt-5.1-codex",
6262
"gpt-5.1-codex-mini",
63+
"gpt-5.6-sol",
64+
"gpt-5.6-terra",
6365
"gpt-5.5",
6466
"gpt-5.4",
6567
"gemini-pro-latest",
@@ -93,6 +95,7 @@ export const curatedCategoryModelIds: Record<
9395
"kimi-k2.5",
9496
"kimi-k2.6",
9597
"kimi-k2",
98+
"gpt-5.6-sol",
9699
"gpt-5.5",
97100
"gpt-5-chat-latest",
98101
"gemini-pro-latest",
@@ -109,6 +112,8 @@ export const curatedCategoryModelIds: Record<
109112
"gemini-3.1-pro-preview",
110113
"gemini-2.5-flash",
111114
"gemini-2.5-flash-lite",
115+
"gpt-5.6-terra",
116+
"gpt-5.6-luna",
112117
"gpt-5.4",
113118
"gpt-5.4-mini",
114119
"claude-sonnet-5",
@@ -123,6 +128,7 @@ export const curatedCategoryModelIds: Record<
123128
"kimi-k2.6",
124129
]),
125130
math: new Set([
131+
"gpt-5.6-sol",
126132
"gpt-5.5-pro",
127133
"gpt-5.5",
128134
"gpt-5.4-pro",

‎packages/models/src/models/openai.ts‎

Lines changed: 201 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1725,6 +1725,207 @@ export const openaiModels = [
17251725
},
17261726
],
17271727
},
1728+
{
1729+
id: "gpt-5.6-sol",
1730+
name: "GPT-5.6 Sol",
1731+
description:
1732+
"Frontier GPT-5.6 model for complex professional work with the highest reasoning capabilities.",
1733+
family: "openai",
1734+
releasedAt: new Date("2026-07-09"),
1735+
providers: [
1736+
{
1737+
providerId: "openai",
1738+
externalId: "gpt-5.6-sol",
1739+
serviceTiers: ["flex", "priority"],
1740+
serviceTierMultipliers: { priority: 2 },
1741+
inputPrice: "5.0e-6",
1742+
outputPrice: "30.0e-6",
1743+
cachedInputPrice: "0.5e-6",
1744+
requestPrice: "0",
1745+
contextSize: 1050000,
1746+
maxOutput: 128000,
1747+
streaming: true,
1748+
vision: true,
1749+
tools: true,
1750+
webSearch: true,
1751+
webSearchPrice: "0.01",
1752+
reasoning: true,
1753+
reasoningOutput: "omit",
1754+
supportsResponsesApi: true,
1755+
jsonOutputSchema: true,
1756+
supportedParameters: [
1757+
"temperature",
1758+
"top_p",
1759+
"frequency_penalty",
1760+
"presence_penalty",
1761+
"response_format",
1762+
],
1763+
jsonOutput: true,
1764+
},
1765+
{
1766+
test: "skip",
1767+
providerId: "azure",
1768+
externalId: "gpt-5.6-sol",
1769+
inputPrice: "5.0e-6",
1770+
outputPrice: "30.0e-6",
1771+
cachedInputPrice: "0.5e-6",
1772+
requestPrice: "0",
1773+
contextSize: 1050000,
1774+
maxOutput: 128000,
1775+
streaming: true,
1776+
vision: true,
1777+
tools: true,
1778+
webSearch: true,
1779+
webSearchPrice: "0.01",
1780+
reasoning: true,
1781+
reasoningOutput: "omit",
1782+
supportsResponsesApi: true,
1783+
jsonOutputSchema: true,
1784+
supportedParameters: [
1785+
"temperature",
1786+
"top_p",
1787+
"frequency_penalty",
1788+
"presence_penalty",
1789+
"response_format",
1790+
],
1791+
jsonOutput: true,
1792+
},
1793+
],
1794+
},
1795+
{
1796+
id: "gpt-5.6-terra",
1797+
name: "GPT-5.6 Terra",
1798+
description:
1799+
"GPT-5.6 model that balances intelligence and cost for everyday workloads.",
1800+
family: "openai",
1801+
releasedAt: new Date("2026-07-09"),
1802+
providers: [
1803+
{
1804+
providerId: "openai",
1805+
externalId: "gpt-5.6-terra",
1806+
serviceTiers: ["flex", "priority"],
1807+
serviceTierMultipliers: { priority: 2 },
1808+
inputPrice: "2.5e-6",
1809+
outputPrice: "15.0e-6",
1810+
cachedInputPrice: "0.25e-6",
1811+
requestPrice: "0",
1812+
contextSize: 1050000,
1813+
maxOutput: 128000,
1814+
streaming: true,
1815+
vision: true,
1816+
tools: true,
1817+
webSearch: true,
1818+
webSearchPrice: "0.01",
1819+
reasoning: true,
1820+
reasoningOutput: "omit",
1821+
supportsResponsesApi: true,
1822+
jsonOutputSchema: true,
1823+
supportedParameters: [
1824+
"temperature",
1825+
"top_p",
1826+
"frequency_penalty",
1827+
"presence_penalty",
1828+
"response_format",
1829+
],
1830+
jsonOutput: true,
1831+
},
1832+
{
1833+
test: "skip",
1834+
providerId: "azure",
1835+
externalId: "gpt-5.6-terra",
1836+
inputPrice: "2.5e-6",
1837+
outputPrice: "15.0e-6",
1838+
cachedInputPrice: "0.25e-6",
1839+
requestPrice: "0",
1840+
contextSize: 1050000,
1841+
maxOutput: 128000,
1842+
streaming: true,
1843+
vision: true,
1844+
tools: true,
1845+
webSearch: true,
1846+
webSearchPrice: "0.01",
1847+
reasoning: true,
1848+
reasoningOutput: "omit",
1849+
supportsResponsesApi: true,
1850+
jsonOutputSchema: true,
1851+
supportedParameters: [
1852+
"temperature",
1853+
"top_p",
1854+
"frequency_penalty",
1855+
"presence_penalty",
1856+
"response_format",
1857+
],
1858+
jsonOutput: true,
1859+
},
1860+
],
1861+
},
1862+
{
1863+
id: "gpt-5.6-luna",
1864+
name: "GPT-5.6 Luna",
1865+
description:
1866+
"Cost-optimized GPT-5.6 model for high-volume tasks like classification and data extraction.",
1867+
family: "openai",
1868+
releasedAt: new Date("2026-07-09"),
1869+
providers: [
1870+
{
1871+
providerId: "openai",
1872+
externalId: "gpt-5.6-luna",
1873+
serviceTiers: ["flex", "priority"],
1874+
serviceTierMultipliers: { priority: 2 },
1875+
inputPrice: "1.0e-6",
1876+
outputPrice: "6.0e-6",
1877+
cachedInputPrice: "0.1e-6",
1878+
requestPrice: "0",
1879+
contextSize: 1050000,
1880+
maxOutput: 128000,
1881+
streaming: true,
1882+
vision: true,
1883+
tools: true,
1884+
webSearch: true,
1885+
webSearchPrice: "0.01",
1886+
reasoning: true,
1887+
reasoningOutput: "omit",
1888+
supportsResponsesApi: true,
1889+
jsonOutputSchema: true,
1890+
supportedParameters: [
1891+
"temperature",
1892+
"top_p",
1893+
"frequency_penalty",
1894+
"presence_penalty",
1895+
"response_format",
1896+
],
1897+
jsonOutput: true,
1898+
},
1899+
{
1900+
test: "skip",
1901+
providerId: "azure",
1902+
externalId: "gpt-5.6-luna",
1903+
inputPrice: "1.0e-6",
1904+
outputPrice: "6.0e-6",
1905+
cachedInputPrice: "0.1e-6",
1906+
requestPrice: "0",
1907+
contextSize: 1050000,
1908+
maxOutput: 128000,
1909+
streaming: true,
1910+
vision: true,
1911+
tools: true,
1912+
webSearch: true,
1913+
webSearchPrice: "0.01",
1914+
reasoning: true,
1915+
reasoningOutput: "omit",
1916+
supportsResponsesApi: true,
1917+
jsonOutputSchema: true,
1918+
supportedParameters: [
1919+
"temperature",
1920+
"top_p",
1921+
"frequency_penalty",
1922+
"presence_penalty",
1923+
"response_format",
1924+
],
1925+
jsonOutput: true,
1926+
},
1927+
],
1928+
},
17281929
{
17291930
id: "gpt-5.2-codex",
17301931
name: "GPT-5.2 Codex",

0 commit comments

Comments
 (0)