BofAI · Will-Guan · Aug 26, 2026 · Aug 26, 2026
diff --git a/docs/llmservice/image-models/gpt-image-2.md b/docs/llmservice/image-models/gpt-image-2.md
@@ -0,0 +1,51 @@
+# GPT-Image-2
+
+## Overview
+
+GPT-Image-2 is an OpenAI image generation and editing model. It accepts text prompts and reference images, and is designed for creative production, marketing assets, and high-quality visual content.
+
+### Billing Method
+
+GPT-Image-2 is billed by usage across text input, cached text input, image input, cached image input, and image output tokens. It does not have a single fixed price per generated image. Resolution, quality, aspect ratio, reference images, and output count can affect the final token usage.
+
+## Key Features
+
+- **Flexible aspect ratios:** Supports auto, 16:9, 3:2, 1:1, 2:3, and 9:16.
+- **Multiple output specifications:** Supports 1K, 2K, and 4K resolutions with Auto, low, medium, and high quality levels.
+- **Batch generation:** Generates 1 to 8 images in a single request.
+- **Reference-image editing:** Accepts up to 16 reference images for high-fidelity editing.
+- **Cost control:** Resolution, quality, and output count can be adjusted for different production requirements.
+
+## Best Use Cases
+
+- Marketing graphics, posters, banners, covers, and social media assets
+- Product concepts, character designs, and visual style exploration
+- High-resolution creative assets that require detailed rendering
+- Image editing based on one or more reference images
+- Batch generation of multiple candidates for comparison
+
+## Capabilities and Limits
+
+| Capability | Description |
+| :--- | :--- |
+| **Aspect ratio** | auto / 16:9 / 3:2 / 1:1 / 2:3 / 9:16 |
+| **Resolution** | 1K / 2K / 4K |
+| **Quality** | Auto / low / medium / high |
+| **Images per request** | 1-8 |
+| **Reference images** | Up to 16 |
+| **Reference-image requirements** | Up to 10 MB per image; jpg, jpeg, and png; URL input |
+| **Transparent background** | Not supported by the current B.AI interface for this model |
+
+## Pricing
+
+Prices are based on token usage. Under the platform-wide conversion of `1 USD = 1,000,000 Credits`, a provider reference price of `USD X / 1M Tokens` corresponds numerically to `X Credits/Token`.
+
+| Billing item | Reference price (USD/1M Tokens) | B.AI rate (Credits/Token) | How it is billed |
+| :--- | ---: | ---: | :--- |
+| **Text input** | `5.00` | `5.00` | Based on text input tokens |
+| **Cached text input** | `1.25` | `1.25` | Applied when cached text input is reused |
+| **Reference-image input** | `8.00` | `8.00` | Based on image input tokens; multiple images accumulate usage |
+| **Cached image input** | `2.00` | `2.00` | Applied when cached image input is reused |
+| **Image output** | `30.00` | `30.00` | Based on generated image output tokens |
+
+**Pricing note:** Documented prices are B.AI standard reference prices for basic billing guidance. B.AI may provide a lower effective usage cost through top-up rewards or account benefits. Prices, bonus Credits, account benefits, and final settlement are subject to the platform display and billing records.
diff --git a/docs/llmservice/image-models/grok-imagine-image-2-0.md b/docs/llmservice/image-models/grok-imagine-image-2-0.md
@@ -0,0 +1,50 @@
+# Grok Imagine Image 2.0
+
+## Overview
+
+Grok Imagine Image 2.0 is a SpaceXAI image generation and editing model. It supports text-to-image generation and multi-image editing, with multiple aspect ratios, 1K and 2K output, and low and medium quality levels.
+
+### Billing Method
+
+Grok Imagine Image 2.0 is billed by the number of images generated. The output price depends on resolution and quality, and each reference image used for editing is billed separately. This model does not use the Input / Cache Write / Cache Read / Output token structure of the text-output model table.
+
+## Key Features
+
+- **Flexible aspect ratios:** Supports auto, 21:9, 16:9, 3:2, 4:3, 1:1, 3:4, 2:3, and 9:16.
+- **Multiple output specifications:** Supports 1K and 2K resolutions with low and medium quality levels. The default quality is medium.
+- **Batch generation:** Generates multiple images from the same prompt.
+- **Multi-image editing:** Accepts up to 3 reference images.
+- **Predictable per-image pricing:** Charges by the actual number of generated images, resolution, and quality.
+
+## Best Use Cases
+
+- Posters, banners, covers, thumbnails, and social media assets
+- Character, environment, product, and visual-style concepts
+- Assets adapted for widescreen, mobile, portrait, and photography layouts
+- Style transfer, content editing, and composition using reference images
+- Batch generation of multiple creative candidates
+
+## Capabilities and Limits
+
+| Capability | Description |
+| :--- | :--- |
+| **Aspect ratio** | auto / 21:9 / 16:9 / 3:2 / 4:3 / 1:1 / 3:4 / 2:3 / 9:16 |
+| **Resolution** | 1K / 2K |
+| **Quality** | low / medium; medium by default |
+| **Batch generation** | Supports multiple outputs from one prompt |
+| **Reference images** | Up to 3 |
+| **Reference-image requirements** | Up to 10 MB per image; jpg, jpeg, and png; URL input |
+| **Output format** | Temporary URL by default; base64 can also be requested |
+
+## Pricing
+
+| Resolution | Quality | Reference price per image | B.AI rate per generated image |
+| :--- | :--- | ---: | ---: |
+| **1K** | low | `$0.04` | **40,000 Credits** |
+| **2K** | low | `$0.06` | **60,000 Credits** |
+| **1K** | medium | `$0.06` | **60,000 Credits** |
+| **2K** | medium | `$0.08` | **80,000 Credits** |
+
+Reference-image input is billed at `$0.01` or **10,000 Credits per image**. For batch generation, output Credits accumulate based on the actual number of generated images.
+
+**Pricing note:** Documented prices are B.AI standard reference prices for basic billing guidance. B.AI may provide a lower effective usage cost through top-up rewards or account benefits. Prices, bonus Credits, account benefits, and final settlement are subject to the platform display and billing records.
diff --git a/docs/llmservice/image-models/pricing.md b/docs/llmservice/image-models/pricing.md
@@ -0,0 +1,40 @@
+# Image Generation Model Pricing
+
+Image generation models may use different billing methods. Some models are billed by text, image input, and image output tokens, while others are billed by each generated image. Check the billing unit shown for each model before estimating cost.
+
+Under the platform-wide conversion of `1 USD = 1,000,000 Credits`, `USD X / 1M Tokens` corresponds numerically to `X Credits/Token` for token-based pricing.
+
+## GPT-Image-2
+
+**Billing method:** Token-based
+
+GPT-Image-2 is billed by actual text input, cached text input, image input, cached image input, and image output token usage. It does not have a single fixed price per generated image. Resolution, quality, aspect ratio, reference images, and output count can affect final usage.
+
+| Billing item | Rate (Credits/Token) |
+| :--- | ---: |
+| Text input | `5.00` |
+| Cached text input | `1.25` |
+| Image input | `8.00` |
+| Cached image input | `2.00` |
+| Image output | `30.00` |
+
+[View GPT-Image-2 model details](./gpt-image-2.md)
+
+## Grok Imagine Image 2.0
+
+**Billing method:** Per generated image
+
+Grok Imagine Image 2.0 is billed by the actual number of generated images. The output rate depends on resolution and quality.
+
+| Resolution | Quality | Output (Credits/Image) |
+| :--- | :--- | ---: |
+| 1K | low | `40,000` |
+| 2K | low | `60,000` |
+| 1K | medium | `60,000` |
+| 2K | medium | `80,000` |
+
+Reference-image input is billed at `10,000 Credits/Image`. For batch generation, output Credits accumulate based on the actual number of generated images.
+
+[View Grok Imagine Image 2.0 model details](./grok-imagine-image-2-0.md)
+
+**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, account benefits, and final settlement are subject to the platform display and billing records.
diff --git a/docs/llmservice/models/deepseek-v4-flash.md b/docs/llmservice/models/deepseek-v4-flash.md
@@ -4,7 +4,7 @@ import ActivityCard from '@site/src/components/ActivityCard';
 
 ## Overview
 
-DeepSeek-V4-Flash is DeepSeek's high-efficiency open-source language model, released alongside V4-Pro on April 24, 2026 under the MIT License. With 284 billion total parameters and only 13 billion active parameters, it delivers performance within striking distance of V4-Pro at roughly one-third of the standard input and output price, making it one of the most cost-effective models available.
+DeepSeek-V4-Flash is DeepSeek's high-efficiency open-source language model, released alongside V4-Pro on April 24, 2026 under the MIT License. With 284 billion total parameters and only 13 billion active parameters, it delivers performance within striking distance of V4-Pro at roughly 3.1x lower cost, making it one of the most cost-effective models available.
 
 <ActivityCard
   variant="free"
@@ -60,5 +60,5 @@ After the offer ends, the model will return to standard pricing. Offer end time,
 | **Busy** | `0.44` | `0.44` | `0.0147` | `1.32` | `-` | Cache Write: `1x` input; Cache Read: `0.0333x` input |
 
 :::info Pricing note
-The table shows the time-based standard reference price for DeepSeek-V4-Flash. Its current limited-time offer applies `0 Credits` to all B.AI Chat and API usage. After the offer ends, the applicable Idle or Busy period, final settlement price, and billing records are subject to the platform display. B.AI may provide lower actual usage costs through top-up bonuses and account benefits.
+The table shows the time-based standard reference price for DeepSeek-V4-Flash. Under DeepSeek's updated rule, from 00:00 UTC+8 on August 23, 2026, Saturdays and Sundays are billed at Idle rates all day. Its current limited-time offer applies `0 Credits` to all B.AI Chat and API usage. After the offer ends, the applicable weekday Idle or Busy period, final settlement price, and billing records are subject to the platform display. B.AI may provide lower actual usage costs through top-up bonuses and account benefits.
 :::
diff --git a/docs/llmservice/models/deepseek-v4-pro.md b/docs/llmservice/models/deepseek-v4-pro.md
@@ -44,5 +44,5 @@ DeepSeek-V4-Pro is DeepSeek's flagship open-source large language model, release
 | **Busy** | `1.32` | `1.32` | `0.044` | `3.96` | `-` | Cache Write: `1x` input; Cache Read: `0.0333x` input |
 
 :::info Pricing note
-DeepSeek V4 Pro uses time-based standard reference pricing. The applicable Idle or Busy period, final settlement price, and billing records are subject to the platform display. B.AI may provide lower actual usage costs through top-up bonuses and account benefits.
+DeepSeek V4 Pro uses time-based standard reference pricing. Under DeepSeek's updated rule, from 00:00 UTC+8 on August 23, 2026, Saturdays and Sundays are billed at Idle rates all day. The applicable weekday Idle or Busy period, final settlement price, and billing records are subject to the platform display. B.AI may provide lower actual usage costs through top-up bonuses and account benefits.
 :::
diff --git a/docs/llmservice/models/gpt-5-6-sol.md b/docs/llmservice/models/gpt-5-6-sol.md
@@ -42,8 +42,7 @@ GPT-5.6 Sol is OpenAI's flagship GPT-5.6 tier, made generally available on July
 
 | 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 |
+| Short context | `4.00` | `5.00` | `0.40` | `20.00` | `10,000` | Standard GPT-5.6 Sol pricing |
 
 :::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.

diff --git a/docs/llmservice/models/mimo-v2.5.md b/docs/llmservice/models/mimo-v2.5.md
@@ -1,9 +1,25 @@
+import ActivityCard from '@site/src/components/ActivityCard';
+
 # MiMo-V2.5
 
 ## Overview
 
 MiMo-V2.5 is an open-weight native omnimodal model from Xiaomi MiMo that entered public beta on April 23, 2026. It combines a 310-billion-parameter sparse Mixture-of-Experts (MoE) language backbone with dedicated vision and audio encoders to understand text, images, video, and audio in long-context agent workflows.
 
+<ActivityCard
+  variant="free"
+  title="MiMo-V2.5"
+  status="Free Offer"
+  detail="API Free · Chat Aug 25"
+>
+MiMo-V2.5 free access is available in phases:
+
+* **API:** MiMo-V2.5 API usage is billed at `0 Credits` from August 24, 2026.
+* **Chat:** The free-access date is August 25, 2026. The exact availability time is subject to actual activation. Once available, MiMo-V2.5 usage is billed at `0 Credits`.
+
+After the offer ends, MiMo-V2.5 will return to standard pricing. The table below shows its standard reference price.
+</ActivityCard>
+
 ## Key Features
 
 * **Native Omnimodal Understanding:** Processes text, image, video, and audio input in one model, using a 729M-parameter vision encoder and a 261M-parameter audio encoder in addition to the language backbone.

diff --git a/docs/llmservice/pricing-and-usage.md b/docs/llmservice/pricing-and-usage.md
@@ -12,7 +12,11 @@ The platform uses a unified Credits system to measure and settle usage across al
 
 **Token usage details:** The response details panel shows a breakdown of token usage, helping you understand where Credits are spent and optimize future usage.
 
-**Model pricing:** Different AI models have different pricing based on their capabilities and compute cost. In general, more capable models consume more Credits. Cache-enabled requests may incur separate cache write and cache read usage. Web search incurs an additional per-use charge. Some models do not support web search and are marked with `-`. See the table below for detailed pricing:
+**Model pricing:** Different AI models use different billing structures based on their capabilities and output type. Text-output models are generally billed by input and output tokens, while image generation models may be billed by token usage or by each generated image. Cache-enabled requests may incur separate cache write and cache read usage. Web search incurs an additional per-use charge. See the sections below for detailed pricing.
+
+### Text-Output Model Pricing
+
+**Image generation pricing:** Image generation models may be billed by token usage or by each generated image. See [Image Generation Model Pricing](./image-models/pricing.md) for the applicable billing unit and rates.
 
 :::info Current promotions and pricing updates
 The table below lists standard reference prices only. For current limited-time offers and pricing updates, see [Promotions and Pricing Updates](./promotions-and-pricing-notices.md). Eligibility, actual settlement prices, and final billing are subject to the platform display.
@@ -37,10 +41,11 @@ The table below lists standard reference prices only. For current limited-time o
 | GLM-5.1           |                  1.40 |                        1.40 |                       0.28 |                   4.40 |                        - |
 | DeepSeek V3.2     |                  0.29 |                        0.29 |                      0.145 |                   0.44 |                        - |
 | DeepSeek-V4-Flash | Idle `0.22`<br/>Busy `0.44` | Idle `0.22`<br/>Busy `0.44` | Idle `0.0073`<br/>Busy `0.0147` | Idle `0.66`<br/>Busy `1.32` | - |
+| DeepSeek-V4-Flash-Vision-Exp | Idle `0.22`<br/>Busy `0.44` | Idle `0.22`<br/>Busy `0.44` | Idle `0.0073`<br/>Busy `0.0147` | Idle `0.66`<br/>Busy `1.32` | - |
 | DeepSeek V4 Pro   | Idle `0.66`<br/>Busy `1.32` | Idle `0.66`<br/>Busy `1.32` | Idle `0.022`<br/>Busy `0.044` | Idle `1.98`<br/>Busy `3.96` | - |
 | Grok 4.6          |                  2.00 |                        2.00 |                       0.50 |                   6.00 |                        - |
 | Grok 4.5          |                  2.00 |                        2.00 |                       0.30 |                   6.00 |                        - |
-| GPT-5.6 Sol       |                  5.00 |                        6.25 |                       0.50 |                  30.00 |                   10,000 |
+| GPT-5.6 Sol       |                  4.00 |                        5.00 |                       0.40 |                  20.00 |                   10,000 |
 | GPT-5.6 Terra     |                  2.00 |                        2.50 |                       0.20 |                  12.00 |                   10,000 |
 | GPT-5.6 Luna      |                  0.20 |                        0.25 |                       0.02 |                   1.20 |                   10,000 |
 | GPT-5.4           |                  2.50 |                        2.50 |                       0.25 |                  15.00 |                   10,000 |
@@ -69,7 +74,7 @@ The table below lists standard reference prices only. For current limited-time o
 | 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. DeepSeek V4 Pro and DeepSeek-V4-Flash use time-based pricing, so their rows list both **Idle** and **Busy** rates; the applicable period and final billing are subject to the platform display. DeepSeek-V4-Flash is currently free on B.AI Chat and API under its limited-time offer, while its row shows the standard reference price. 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.
+The main pricing table shows the currently effective standard reference price for each model. DeepSeek V4 Pro, DeepSeek-V4-Flash, and DeepSeek-V4-Flash-Vision-Exp use time-based pricing, so their rows list both **Idle** and **Busy** rates. Under DeepSeek's updated rule, from 00:00 UTC+8 on August 23, 2026, Saturdays and Sundays are billed at Idle rates all day; the applicable weekday period and final billing are subject to the platform display. DeepSeek-V4-Flash is currently free on B.AI Chat and API, while DeepSeek-V4-Flash-Vision-Exp is currently free for B.AI API use; both rows show standard reference prices. MiMo-V2.5 API usage is currently free, and its B.AI Chat free-access date is August 25, 2026; the exact availability time is subject to actual activation. Its row shows the standard reference price. 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