> ## Documentation Index
> Fetch the complete documentation index at: https://docs.savegate.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Supported Models
> Complete list of AI models available through SaveGate
## Overview
SaveGate provides access to **208+ AI models** from leading providers. All models are accessible through a unified API using the same authentication and request format.
Pricing shown is per 1M tokens (input/output).
## OpenAI Models
### Chat Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| --------------------------------------- | --------- | ---------- | -------- | --------- |
| `gpt-5.4` | 1.1M | 128K | \$2.50 | \$15 |
| `gpt-5.4-2026-03-05` | 1.1M | 128K | \$2.50 | \$15 |
| `gpt-5.4-mini` | 272K | 128K | \$0.25 | \$2 |
| `gpt-5.4-nano` | 272K | 128K | \$0.05 | \$0.40 |
| `gpt-5.3-chat-latest` | 128K | 16.4K | \$1.75 | \$14 |
| `gpt-5.2` | 272K | 128K | \$1.75 | \$14 |
| `gpt-5.2-chat-latest` | 128K | 16.4K | \$1.75 | \$14 |
| `gpt-5.2-2025-12-11` | 272K | 128K | \$1.75 | \$14 |
| `gpt-5.1` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5.1-chat-latest` | 128K | 16.4K | \$1.25 | \$10 |
| `gpt-5.1-2025-11-13` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5-chat` | 128K | 16.4K | \$1.25 | \$10 |
| `gpt-5-chat-latest` | 128K | 16.4K | \$1.25 | \$10 |
| `gpt-5-search-api` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5-2025-08-07` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5-search-api-2025-10-14` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5-mini` | 272K | 128K | \$0.25 | \$2 |
| `gpt-5-mini-2025-08-07` | 272K | 128K | \$0.25 | \$2 |
| `gpt-5-nano` | 272K | 128K | \$0.05 | \$0.40 |
| `gpt-5-nano-2025-08-07` | 272K | 128K | \$0.05 | \$0.40 |
| `o4-mini` | 200K | 100K | \$1.10 | \$4.40 |
| `o4-mini-2025-04-16` | 200K | 100K | \$1.10 | \$4.40 |
| `chatgpt-4o-latest` | 128K | 4.1K | \$5 | \$15 |
| `gpt-4o` | 128K | 16.4K | \$2.50 | \$10 |
| `gpt-4o-search-preview` | 128K | 16.4K | \$2.50 | \$10 |
| `gpt-4o-2024-05-13` | 128K | 4.1K | \$5 | \$15 |
| `gpt-4o-2024-08-06` | 128K | 16.4K | \$2.50 | \$10 |
| `gpt-4o-2024-11-20` | 128K | 16.4K | \$2.50 | \$10 |
| `gpt-4o-search-preview-2025-03-11` | 128K | 16.4K | \$2.50 | \$10 |
| `gpt-4o-mini` | 128K | 16.4K | \$0.15 | \$0.60 |
| `gpt-4o-mini-search-preview` | 128K | 16.4K | \$0.15 | \$0.60 |
| `gpt-4o-mini-2024-07-18` | 128K | 16.4K | \$0.15 | \$0.60 |
| `gpt-4o-mini-search-preview-2025-03-11` | 128K | 16.4K | \$0.15 | \$0.60 |
| `gpt-4-turbo` | 128K | 4.1K | \$10 | \$30 |
| `gpt-4.1` | 1.0M | 32.8K | \$2 | \$8 |
| `gpt-4-turbo-2024-04-09` | 128K | 4.1K | \$10 | \$30 |
| `gpt-4.1-2025-04-14` | 1.0M | 32.8K | \$2 | \$8 |
| `gpt-4.1-mini` | 1.0M | 32.8K | \$0.40 | \$1.60 |
| `gpt-4.1-mini-2025-04-14` | 1.0M | 32.8K | \$0.40 | \$1.60 |
| `gpt-4.1-nano` | 1.0M | 32.8K | \$0.10 | \$0.40 |
| `gpt-4.1-nano-2025-04-14` | 1.0M | 32.8K | \$0.10 | \$0.40 |
| `gpt-4` | 8.2K | 4.1K | \$30 | \$60 |
| `gpt-4-0613` | 8.2K | 4.1K | \$30 | \$60 |
| `o3` | 200K | 100K | \$2 | \$8 |
| `o3-2025-04-16` | 200K | 100K | \$2 | \$8 |
| `o3-mini` | 200K | 100K | \$1.10 | \$4.40 |
| `o3-mini-2025-01-31` | 200K | 100K | \$1.10 | \$4.40 |
| `gpt-3.5-turbo` | 16.4K | 4.1K | \$0.50 | \$1.50 |
| `o1` | 200K | 100K | \$15 | \$60 |
| `o1-2024-12-17` | 200K | 100K | \$15 | \$60 |
| `gpt-realtime` | 32K | 4.1K | \$4 | \$16 |
| `gpt-realtime-1.5` | 32K | 4.1K | \$4 | \$16 |
| `gpt-realtime-2025-08-28` | 32K | 4.1K | \$4 | \$16 |
| `gpt-realtime-mini` | 128K | 4.1K | \$0.60 | \$2.40 |
| `gpt-realtime-mini-2025-10-06` | 128K | 4.1K | \$0.60 | \$2.40 |
| `gpt-realtime-mini-2025-12-15` | 128K | 4.1K | \$0.60 | \$2.40 |
### Responses Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ---------------------------------- | --------- | ---------- | -------- | --------- |
| `gpt-5.4-pro` | 1.1M | 128K | \$30 | \$180 |
| `gpt-5.4-pro-2026-03-05` | 1.1M | 128K | \$30 | \$180 |
| `gpt-5.3-codex` | 272K | 128K | \$1.75 | \$14 |
| `gpt-5.2-pro` | 272K | 128K | \$21 | \$168 |
| `gpt-5.2-pro-2025-12-11` | 272K | 128K | \$21 | \$168 |
| `gpt-5.2-codex` | 272K | 128K | \$1.75 | \$14 |
| `gpt-5.1-codex` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5.1-codex-max` | 272K | 128K | \$1.25 | \$10 |
| `gpt-5.1-codex-mini` | 272K | 128K | \$0.25 | \$2 |
| `gpt-5-pro` | 128K | 272K | \$15 | \$120 |
| `gpt-5-pro-2025-10-06` | 128K | 272K | \$15 | \$120 |
| `gpt-5-codex` | 272K | 128K | \$1.25 | \$10 |
| `o4-mini-deep-research` | 200K | 100K | \$2 | \$8 |
| `o4-mini-deep-research-2025-06-26` | 200K | 100K | \$2 | \$8 |
| `o3-pro` | 200K | 100K | \$20 | \$80 |
| `o3-pro-2025-06-10` | 200K | 100K | \$20 | \$80 |
| `o3-deep-research` | 200K | 100K | \$10 | \$40 |
| `o3-deep-research-2025-06-26` | 200K | 100K | \$10 | \$40 |
| `o1-pro` | 200K | 100K | \$150 | \$600 |
| `o1-pro-2025-03-19` | 200K | 100K | \$150 | \$600 |
| `codex-mini-latest` | 200K | 100K | \$1.50 | \$6 |
### Embedding Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| --------------------------- | --------- | ---------- | -------- | --------- |
| `text-embedding-3-large` | 8.2K | - | \$0.13 | - |
| `text-embedding-3-small` | 8.2K | - | \$0.02 | - |
| `text-embedding-ada-002` | 8.2K | - | \$0.10 | - |
| `text-embedding-ada-002-v2` | 8.2K | - | \$0.10 | - |
### Image Generation Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ------------------------------- | --------- | ---------- | -------- | --------- |
| `chatgpt-image-latest` | - | - | \$5 | - |
| `standard/1024-x-1024/dall-e-3` | - | - | - | - |
| `standard/1024-x-1792/dall-e-3` | - | - | - | - |
| `standard/1792-x-1024/dall-e-3` | - | - | - | - |
| `gpt-image-1` | - | - | \$5 | - |
| `gpt-image-1.5` | - | - | \$5 | \$10 |
| `gpt-image-1.5-2025-12-16` | - | - | \$5 | \$10 |
| `gpt-image-1-mini` | - | - | \$2 | - |
Image generation models also support resolution and quality variants (e.g., `hd/1024-x-1024/dall-e-3`). Use the `/v1/models` API endpoint to see all available variants.
### Audio Speech Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ---------------------------- | --------- | ---------- | -------- | --------- |
| `gpt-4o-mini-tts` | - | - | \$2.50 | \$10 |
| `gpt-4o-mini-tts-2025-03-20` | - | - | \$2.50 | \$10 |
| `gpt-4o-mini-tts-2025-12-15` | - | - | \$2.50 | \$10 |
| `tts-1` | - | - | - | - |
| `tts-1-1106` | - | - | - | - |
| `tts-1-hd` | - | - | - | - |
| `tts-1-hd-1106` | - | - | - | - |
### Audio Transcription Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ----------------------------------- | --------- | ---------- | -------- | --------- |
| `gpt-4o-transcribe` | 16K | 2K | \$2.50 | \$10 |
| `gpt-4o-transcribe-diarize` | 16K | 2K | \$2.50 | \$10 |
| `gpt-4o-mini-transcribe` | 16K | 2K | \$1.25 | \$5 |
| `gpt-4o-mini-transcribe-2025-03-20` | 16K | 2K | \$1.25 | \$5 |
| `gpt-4o-mini-transcribe-2025-12-15` | 16K | 2K | \$1.25 | \$5 |
| `whisper-1` | - | - | - | - |
### Video Generation Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| --------------------- | --------- | ---------- | -------- | --------- |
| `sora-2-pro` | - | - | - | - |
| `sora-2-pro-high-res` | - | - | - | - |
| `sora-2` | - | - | - | - |
### Completion Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ------------- | --------- | ---------- | -------- | --------- |
| `babbage-002` | 16.4K | 4.1K | \$0.40 | \$0.40 |
### Moderation Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ---------------------------- | --------- | ---------- | -------- | --------- |
| `omni-moderation-latest` | 32.8K | 0 | - | - |
| `text-moderation-007` | 32.8K | 0 | - | - |
| `text-moderation-latest` | 32.8K | 0 | - | - |
| `text-moderation-stable` | 32.8K | 0 | - | - |
| `omni-moderation-2024-09-26` | 32.8K | 0 | - | - |
## Anthropic Claude Models
### Chat Models
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ---------------------------- | --------- | ---------- | -------- | --------- |
| `claude-opus-4-7` | 1M | 128K | \$5 | \$25 |
| `claude-opus-4-6` | 1M | 128K | \$5 | \$25 |
| `claude-sonnet-4-6` | 1M | 64K | \$3 | \$15 |
| `claude-opus-4-5` | 200K | 64K | \$5 | \$25 |
| `claude-opus-4-5-20251101` | 200K | 64K | \$5 | \$25 |
| `claude-sonnet-4-5` | 200K | 64K | \$3 | \$15 |
| `claude-sonnet-4-5-20250929` | 200K | 64K | \$3 | \$15 |
| `claude-haiku-4-5` | 200K | 64K | \$1 | \$5 |
| `claude-haiku-4-5-20251001` | 200K | 64K | \$1 | \$5 |
| `claude-opus-4-20250514` | 200K | 32K | \$15 | \$75 |
| `claude-sonnet-4-20250514` | 1M | 64K | \$3 | \$15 |
| `claude-opus-4-1` | 200K | 32K | \$15 | \$75 |
| `claude-opus-4-1-20250805` | 200K | 32K | \$15 | \$75 |
| `claude-4-opus-20250514` | 200K | 32K | \$15 | \$75 |
| `claude-4-sonnet-20250514` | 1M | 64K | \$3 | \$15 |
| `claude-3-haiku-20240307` | 200K | 4.1K | \$0.25 | \$1.25 |
## Google Gemini Models
### Chat Models (OpenAI-Compatible)
Chat models are accessed via the OpenAI-compatible endpoint. Use the model IDs below directly — no prefix needed.
| Model ID | Max Input | Max Output | Input/1M | Output/1M |
| ----------------------------- | --------- | ---------- | -------- | --------- |
| `gemini-2.5-pro` | 1.0M | 64K | \$1.88 | \$15 |
| `gemini-2.5-pro-thinking` | 1.0M | 64K | \$1.88 | \$15 |
| `gemini-2.5-pro-nothinking` | 1.0M | 64K | \$1.88 | \$15 |
| `gemini-2.5-flash` | 1.0M | 64K | \$0.45 | \$3.75 |
| `gemini-2.5-flash-nothinking` | 1.0M | 64K | \$0.45 | \$3.75 |
### Image Generation Models (Native Gemini API)
Image generation models use the native Google Gemini API — **not** the OpenAI-compatible endpoint. Use the [Google Gemini SDK](https://ai.google.dev/gemini-api/docs) with your SaveGate API key passed via the `x-goog-api-key` header.
| Model ID | Cost per Image |
| ------------------------------- | -------------- |
| `gemini-3-pro-image-preview-4k` | \$0.30 |
**Example: Generate an image**
```bash theme={null}
curl -s 'https://api.savegate.ai/v1beta/models/gemini-3-pro-image-preview-4k:generateContent' \
-H 'Content-Type: application/json' \
-H 'x-goog-api-key: sg-xxx' \
-d '{
"contents": [
{
"parts": [
{
"text": "A futuristic city at sunset"
}
]
}
]
}' \
| jq -r '.candidates[0].content.parts[0].inlineData.data' \
| base64 --decode > output.png
```
## ElevenLabs Models
### Text-to-Speech Models
| Model ID | Mode | Pricing |
| ---------------------- | ------------ | ---------------------- |
| `elevenlabs/eleven_v3` | Audio Speech | \$0.18 / 1K characters |
### Speech-to-Text Models
| Model ID | Mode | Pricing |
| ----------------------------------- | ------------------- | ------------- |
| `elevenlabs/scribe_v1_experimental` | Audio Transcription | \$0.22 / hour |
## Using Models
Simply specify the model ID in your API request:
```python Python theme={null}
response = client.chat.completions.create(
model="gpt-5.4", # Specify model here
messages=[{"role": "user", "content": "Hello!"}]
)
```
```javascript Node.js theme={null}
const response = await client.chat.completions.create({
model: 'claude-sonnet-4-6', // Specify model here
messages: [{ role: 'user', content: 'Hello!' }]
});
```
## Model Updates
SaveGate automatically updates model versions to the latest stable releases. For version pinning, use specific model IDs (with date suffix) when available.
Use the `/v1/models` API endpoint to get the full list of available models with their capabilities and pricing information in real-time.