Skip to content

Add GPT-5.6 model pricing docs #70

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Will-Guan merged 2 commits into from
Jul 10, 2026
Merged

Add GPT-5.6 model pricing docs #70

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d9c2682
Add GPT-5.6 model pricing docs
Jul 10, 2026
838d63e
update version
jizhen181-dot Jul 10, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 52 additions & 0 deletions docs/llmservice/models/gpt-5-6-luna.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
# GPT-5.6 Luna

## Overview

GPT-5.6 Luna is OpenAI's cost-oriented GPT-5.6 tier, made generally available on July 9, 2026. In the API, its model ID is `gpt-5.6-luna`, and OpenAI positions it for efficient, high-volume workloads.

## Key Features

* **Cost-Oriented GPT-5.6 Tier**: Designed for high-volume workloads where lower token cost and responsiveness are important.
* **Expanded Reasoning Controls**: Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max` reasoning effort; GPT-5.6 also supports Pro mode through `reasoning.mode: "pro"`.
* **Agentic Tooling**: Supports Responses API workflows, functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta.
* **Large Working Context**: Supports a 1.05M-token context window and up to 128K output tokens, with a February 16, 2026 knowledge cutoff.

## Best Use Cases

* **High-Volume Assistance**: Summarization, rewriting, classification, extraction, routing, and lightweight analysis.
* **Responsive Product Features**: User-facing assistants, workflow copilots, and routine automation with predictable cost constraints.
* **First-Pass Technical Work**: Initial issue triage, draft code review, test-case generation, and document analysis before escalating difficult cases to Terra or Sol.

## Capabilities and Limitations

| Capability | Description |
| :------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Reasoning** | Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max`; API Pro mode is enabled with `reasoning.mode: "pro"` rather than a separate Pro model slug. |
| **Coding** | Official benchmark examples include SWE-Bench Pro 62.7%, DeepSWE v1.1 67.2%, and Terminal-Bench 2.1 84.7%. |
| **Creative Writing** | General text generation is supported. |
| **Multimodal** | Text and image input with text output; GPT-5.6 preserves original image dimensions for `original` or `auto` detail. |
| **Response Speed** | Public API latency depends on reasoning effort, tools, and processing tier; Luna is positioned for efficient, high-volume use. |
| **Context Window** | 1.05M tokens. |
| **Max Output** | 128K tokens. |
| **Tool Use** | Functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta. |
| **Multilingual** | OpenAI API model documentation lists multilingual support across current models. |

### Known Limitations

* Luna is positioned below Terra and Sol for capability, so complex reasoning, sensitive technical work, and quality-first workflows may need escalation.
* Higher reasoning efforts, Pro mode, multi-agent runs, and long-context use can increase latency and cost.
* Cyber and biology safeguards may refuse, block, or pause sensitive requests, including some legitimate dual-use work.
* Prompt cache writes are billed on GPT-5.6 models, so cache breakpoints should be monitored with `cache_write_tokens` and `cached_tokens`.

## Credits Usage

| Context | Input (Credits/Token) | Cache Write (Credits/Token) | Cache Read (Credits/Token) | Output (Credits/Token) | Web Search (Credits/Use) | Billing Notes |
| :--- | --------------------: | --------------------------: | -------------------------: | ---------------------: | -----------------------: | :--- |
| Short context | `1.00` | `1.25` | `0.10` | `6.00` | `10,000` | Standard GPT-5.6 Luna pricing |
| Long context (>272K input tokens) | `2.00` | `2.50` | `0.20` | `9.00` | `10,000` | Long-context pricing tier |

:::info Pricing note
Prices shown in the documentation are B.AI standard reference prices for base billing purposes. B.AI may provide lower actual usage costs through top-up bonuses and account benefits. Specific prices, bonus Credits, and account benefits are subject to the platform display and final billing records.
:::

* For GPT-5.6 models, cache writes are billed at 1.25x the uncached input rate; cache reads remain discounted at 0.1x input.
52 changes: 52 additions & 0 deletions docs/llmservice/models/gpt-5-6-sol.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
# GPT-5.6 Sol

## Overview

GPT-5.6 Sol is OpenAI's flagship GPT-5.6 tier, made generally available on July 9, 2026. In the API, its model ID is `gpt-5.6-sol`, and the `gpt-5.6` alias routes to this tier. It is designed for complex professional work, reasoning, coding, cybersecurity, science, and tool-heavy workflows.

## Key Features

* **GPT-5.6 Flagship Tier**: Designed for complex professional work, reasoning, coding, cybersecurity, science, and tool-heavy workflows.
* **Expanded Reasoning Controls**: Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max` reasoning effort; GPT-5.6 also supports Pro mode through `reasoning.mode: "pro"`.
* **Agentic Tooling**: Supports Responses API workflows, functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta.
* **Large Working Context**: Supports a 1.05M-token context window and up to 128K output tokens, with a February 16, 2026 knowledge cutoff.

## Best Use Cases

* **Complex Software Engineering**: Repository-level debugging, command-line workflows, multi-file implementation, validation, and code review.
* **Long-Horizon Knowledge Work**: Browser, document, spreadsheet, presentation, and computer-use tasks that need sustained context and tool coordination.
* **Defensive Security and Research Support**: Authorized vulnerability review, patch development, threat modeling, biology workflow analysis, and scientific research assistance with human oversight.

## Capabilities and Limitations

| Capability | Description |
| :------------------- | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Reasoning** | Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max`; API Pro mode is enabled with `reasoning.mode: "pro"` rather than a separate Pro model slug. |
| **Coding** | Official benchmark examples include SWE-Bench Pro 64.6%, DeepSWE v1.1 72.7%, Terminal-Bench 2.1 88.8%, and Terminal-Bench 2.1 Ultra 91.9%. |
| **Creative Writing** | General text generation is supported. |
| **Multimodal** | Text and image input with text output; GPT-5.6 preserves original image dimensions for `original` or `auto` detail. |
| **Response Speed** | Public API latency depends on reasoning effort, tools, and processing tier; OpenAI also announced Cerebras access for Sol at up to 750 tokens per second for select customers. |
| **Context Window** | 1.05M tokens. |
| **Max Output** | 128K tokens. |
| **Tool Use** | Functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta. |
| **Multilingual** | OpenAI API model documentation lists multilingual support across current models. |

### Known Limitations

* Higher reasoning efforts, Pro mode, multi-agent runs, and long-context use can increase latency and cost.
* Cyber and biology safeguards may refuse, block, or pause sensitive requests, including some legitimate dual-use work.
* OpenAI says GPT-5.6 does not cross the Critical threshold in cybersecurity or biology evaluations, so it should not be treated as an autonomous operator for high-risk work.
* Prompt cache writes are billed on GPT-5.6 models, so cache breakpoints should be monitored with `cache_write_tokens` and `cached_tokens`.

## Credits Usage

| Context | Input (Credits/Token) | Cache Write (Credits/Token) | Cache Read (Credits/Token) | Output (Credits/Token) | Web Search (Credits/Use) | Billing Notes |
| :--- | --------------------: | --------------------------: | -------------------------: | ---------------------: | -----------------------: | :--- |
| Short context | `5.00` | `6.25` | `0.50` | `30.00` | `10,000` | Standard GPT-5.6 Sol pricing |
| Long context (>272K input tokens) | `10.00` | `12.50` | `1.00` | `45.00` | `10,000` | Long-context pricing tier |

:::info Pricing note
Prices shown in the documentation are B.AI standard reference prices for base billing purposes. B.AI may provide lower actual usage costs through top-up bonuses and account benefits. Specific prices, bonus Credits, and account benefits are subject to the platform display and final billing records.
:::

* For GPT-5.6 models, cache writes are billed at 1.25x the uncached input rate; cache reads remain discounted at 0.1x input.
52 changes: 52 additions & 0 deletions docs/llmservice/models/gpt-5-6-terra.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
# GPT-5.6 Terra

## Overview

GPT-5.6 Terra is OpenAI's balanced GPT-5.6 tier, made generally available on July 9, 2026. In the API, its model ID is `gpt-5.6-terra`, and OpenAI positions it for workloads that balance capability and cost.

## Key Features

* **Balanced GPT-5.6 Tier**: Positioned between Sol and Luna for production workflows that need strong reasoning at lower cost than Sol.
* **Expanded Reasoning Controls**: Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max` reasoning effort; GPT-5.6 also supports Pro mode through `reasoning.mode: "pro"`.
* **Agentic Tooling**: Supports Responses API workflows, functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta.
* **Large Working Context**: Supports a 1.05M-token context window and up to 128K output tokens, with a February 16, 2026 knowledge cutoff.

## Best Use Cases

* **Production Knowledge Work**: Drafting, analysis, document processing, structured research, and multi-step business workflows.
* **Software Engineering Assistance**: Debugging, test planning, code review, repository navigation, and implementation support where cost control matters.
* **Scaled Professional Automation**: Tool-heavy workflows that need strong quality but do not require Sol for every request.

## Capabilities and Limitations

| Capability | Description |
| :------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Reasoning** | Supports `none`, `low`, `medium`, `high`, `xhigh`, and `max`; API Pro mode is enabled with `reasoning.mode: "pro"` rather than a separate Pro model slug. |
| **Coding** | Official benchmark examples include SWE-Bench Pro 63.4%, DeepSWE v1.1 69.6%, and Terminal-Bench 2.1 87.4%. |
| **Creative Writing** | General text generation is supported. |
| **Multimodal** | Text and image input with text output; GPT-5.6 preserves original image dimensions for `original` or `auto` detail. |
| **Response Speed** | Public API latency depends on reasoning effort, tools, and processing tier; Terra is positioned as the balanced tier for intelligence and cost. |
| **Context Window** | 1.05M tokens. |
| **Max Output** | 128K tokens. |
| **Tool Use** | Functions, web search, file search, computer use, Programmatic Tool Calling, persisted reasoning, and multi-agent beta. |
| **Multilingual** | OpenAI API model documentation lists multilingual support across current models. |

### Known Limitations

* Terra is positioned below Sol for capability, so the hardest quality-first workflows may need Sol, Pro mode, or higher reasoning effort.
* Higher reasoning efforts, Pro mode, multi-agent runs, and long-context use can increase latency and cost.
* Cyber and biology safeguards may refuse, block, or pause sensitive requests, including some legitimate dual-use work.
* Prompt cache writes are billed on GPT-5.6 models, so cache breakpoints should be monitored with `cache_write_tokens` and `cached_tokens`.

## Credits Usage

| Context | Input (Credits/Token) | Cache Write (Credits/Token) | Cache Read (Credits/Token) | Output (Credits/Token) | Web Search (Credits/Use) | Billing Notes |
| :--- | --------------------: | --------------------------: | -------------------------: | ---------------------: | -----------------------: | :--- |
| Short context | `2.50` | `3.125` | `0.25` | `15.00` | `10,000` | Standard GPT-5.6 Terra pricing |
| Long context (>272K input tokens) | `5.00` | `6.25` | `0.50` | `22.50` | `10,000` | Long-context pricing tier |

:::info Pricing note
Prices shown in the documentation are B.AI standard reference prices for base billing purposes. B.AI may provide lower actual usage costs through top-up bonuses and account benefits. Specific prices, bonus Credits, and account benefits are subject to the platform display and final billing records.
:::

* For GPT-5.6 models, cache writes are billed at 1.25x the uncached input rate; cache reads remain discounted at 0.1x input.
7 changes: 5 additions & 2 deletions docs/llmservice/pricing-and-usage.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,9 @@ The platform uses a unified Credits system to measure and settle usage across al
| DeepSeek V3.2 | 0.29 | 0.29 | 0.145 | 0.44 | - |
| DeepSeek V4 Flash | 0.28 | 0.28 | 0.0056 | 0.56 | - |
| DeepSeek V4 Pro | 0.87 | 0.87 | 0.0087 | 1.74 | - |
| GPT-5.6 Sol | 5.00 | 6.25 | 0.50 | 30.00 | 10,000 |
| GPT-5.6 Terra | 2.50 | 3.125 | 0.25 | 15.00 | 10,000 |
| GPT-5.6 Luna | 1.00 | 1.25 | 0.10 | 6.00 | 10,000 |
| GPT-5.4 | 2.50 | 2.50 | 0.25 | 15.00 | 10,000 |
| GPT-5.5 | 5.00 | 5.00 | 0.50 | 30.00 | 10,000 |
| GPT-5.5 Instant | 5.00 | 5.00 | 0.50 | 30.00 | 10,000 |
Expand All @@ -47,7 +50,7 @@ The platform uses a unified Credits system to measure and settle usage across al
| Gemini 3 Flash | 0.50 | 0.50 | 0.05 | 3.00 | 14,000 |

:::caution Main table scope
The main pricing table shows the currently effective standard reference price for each model. The `Cache Write` column represents the billing rate when cache writing occurs; it does not imply a unified cache TTL across all models. Cache behavior, retention time, and extended caching options may vary by model provider. If a model has special caching rules, 1-hour cache write pricing, or time-based pricing, please refer to the corresponding model detail page.
The main pricing table shows the currently effective standard reference price for each model. The `Cache Write` column represents the billing rate when cache writing occurs; it does not imply a unified cache TTL across all models. Cache behavior, retention time, long-context pricing, and extended caching options may vary by model provider. If a model has special caching rules, long-context pricing, 1-hour cache write pricing, or time-based pricing, please refer to the corresponding model detail page.
:::

:::info Pricing note
Expand All @@ -56,7 +59,7 @@ Prices shown in the documentation are B.AI standard reference prices for base bi

### Cache Pricing Notes

- **Cache Write:** The cost when tokens are first written into the prompt cache. Most providers charge no premium and use the same rate as standard input pricing. Claude models apply a 25% premium on cache writes.
- **Cache Write:** The cost when tokens are first written into the prompt cache. Most providers charge no premium and use the same rate as standard input pricing. Claude models and GPT-5.6 models apply a 25% premium on cache writes; GPT-5.6 long-context cache write rates follow the corresponding long-context input tier.
- **Cache Read:** The discounted cost when cached tokens are reused in subsequent requests.
- **Example with Caching:** If you use Claude Sonnet 4.6 with 1000 tokens cached:
- First request (cache write): `1000 × 3.75 = 3,750 credits`
Expand Down
Loading
Loading