diff --git a/services/site/src/docs/content/console/models.mdx b/services/site/src/docs/content/console/models.mdx index d3d9bfd6e785..e509a966a15a 100644 --- a/services/site/src/docs/content/console/models.mdx +++ b/services/site/src/docs/content/console/models.mdx @@ -116,6 +116,10 @@ translates. See [Base URLs](/console/harnesses#base-urls) for the full set. | Gemini Decision 1 | gemini-decision-1 | `https://opencode.ai/inference/systemone/v1/systemone` | - | | Jev 1.13 | jev-1.13 | `https://opencode.ai/inference/systemone/v1/systemone` | - | | Jev 1.13 Free | jev-1.13-free | `https://opencode.ai/inference/systemone/v1/systemone` | - | +| GPT Image 2.5 Sunburst | gpt-image-2.5-sunburst | `https://opencode.ai/inference/openai/v1/images/generations` | `@ai-sdk/openai` | +| Nano Banana 2.1 | gemini-nano-banana-2.1 | `https://opencode.ai/inference/google/v1beta/models/gemini-nano-banana-2.1:generateContent` | `@ai-sdk/google` | +| Grok Imagine Image 2.0 | grok-imagine-image-2.0 | `https://opencode.ai/inference/xai/v1/images/generations` | `@ai-sdk/xai` | +| GPT-Live 1 | gpt-live-1 | `https://opencode.ai/inference/openai/v1/live/sessions` | - | | Big Pickle | big-pickle | `https://opencode.ai/inference/openai/v1/chat/completions` | `@ai-sdk/openai-compatible` | | Iceman Free | iceman-free | `https://opencode.ai/inference/openai/v1/chat/completions` | `@ai-sdk/openai-compatible` | | Inkling 2 Small Free | inkling-2-small-free | `https://opencode.ai/inference/openai/v1/chat/completions` | `@ai-sdk/openai-compatible` | @@ -221,9 +225,34 @@ curl https://opencode.ai/inference/systemone/v1/systemone \ }' ``` +### Image and voice models + +`gpt-image-2.5-sunburst`, `gemini-nano-banana-2.1`, and `grok-imagine-image-2.0` generate and edit images, and +`gpt-live-1` runs voice sessions. They aren't chat models, so they never appear in `/models` or the models endpoint, +and `opencode/` doesn't select them. Call them on their native API with a Console key, or let OpenCode's +[`experimental_image_generate`](/tools#image-generate) tool use an image model by setting +`experimental.image_generate.model`, for example `opencode/gpt-image-2.5-sunburst`. + +```bash +curl https://opencode.ai/inference/openai/v1/images/generations \ + -H "Authorization: Bearer $OPENCODE_API_KEY" \ + -H "Content-Type: application/json" \ + -d '{ + "model": "gpt-image-2.5-sunburst", + "prompt": "A mountain at sunrise", + "size": "1024x1024" + }' +``` + +Google takes its native `contents` and `generationConfig`, and xAI its native options such as `resolution` and +`quality`. OpenAI Images and xAI Images also accept edits at `/images/edits`. For `gpt-live-1`, start a WebRTC session +at `POST /inference/openai/v1/live/sessions` with OpenAI's session request and return its SDP answer; session time +bills per minute as it runs. + ## Pricing -Console uses pay-as-you-go pricing. Below are the prices **per 1M tokens**. +Console uses pay-as-you-go pricing. Below are the prices **per 1M tokens**, with image and voice models priced per +image or per minute.
@@ -349,6 +378,15 @@ models to generate session titles. A processing fee of 4.6% + $0.31 is added to each credit purchase to cover payment costs; we don't charge anything beyond that. +### Image and voice pricing + +| Model | Price | Notes | +| ---------------------- | --------------------------- | ---------------------------------------------------------------------------------- | +| GPT Image 2.5 Sunburst | $1.00 / image | Plus $8 / $30 per 1M input / output tokens, and $2 per 1M cached input tokens. | +| Nano Banana 2.1 | $1.50 / $7.50 / $30.00 per 1M tokens | Input / text output / image output. About $0.034 per 1K image, $0.050 per 2K, $0.113 per 4K. | +| Grok Imagine Image 2.0 | $0.04 / image | $0.01 per input image; quality and resolution can raise the output price to $0.05–$0.08. | +| GPT-Live 1 | $0.05 / min | Session time, as OpenAI reports each whole minute; 15 seconds minimum. | + ### Free models - GLM-5.3-Flash Free is free on OpenCode for a limited time. Its provider follows a zero-retention policy and does not use your data for model training.