Deep Research Max Preview (Apr-21-2026) pricing, context window and limits
Deep Research Max Preview (Apr-21-2026) costs $2 per million input tokens and $12 per million output tokens at Google Gemini list price. It accepts up to 131,072 tokens of context and returns up to 65,536 tokens.
- Model id
deep-research-max-preview-04-2026 - Snapshot
What does Deep Research Max Preview (Apr-21-2026) cost?
| Input | $2 |
|---|---|
| Output | $12 |
| Cached input (read) | $0.2 |
| Cache write | — |
US dollars per million tokens. A dash means no price is published.
How much can Deep Research Max Preview (Apr-21-2026) read and write?
| Context window | 131,072 |
|---|---|
| Max output | 65,536 |
What can Deep Research Max Preview (Apr-21-2026) do?
| Vision (image input) | Yes |
|---|---|
| Tool calling | Yes |
| Structured output | No |
| Reasoning | Yes |
| Input | text, image, video, audio, pdf |
| Output | text, image |
What does a request to Deep Research Max Preview (Apr-21-2026) cost?
| A typical request: 1,000 tokens in, 500 out | $0.008 |
|---|---|
| A long request: 10,000 tokens in, 1,000 out | $0.032 |
| 1,000 typical requests | $8 |
| 1 million tokens in and 1 million out | $14 |
Input read from the provider's prompt cache costs $0.2 per million tokens instead of $2, 90% less. A request that repeats a 1,000-token prompt from the cache costs $0.0062 instead of $0.008.
Among 34 Google Gemini models with a published price, Deep Research Max Preview (Apr-21-2026) is the 28th cheapest by input price. Its context window is the 20th largest of 39. List prices only; discounts, batch prices and prices above 200K tokens are not shown.
How do I call Deep Research Max Preview (Apr-21-2026) through DVARA?
Configure the gemini provider in the DVARA LLM Gateway and send the model id deep-research-max-preview-04-2026. DVARA, the AI governance platform, applies your policies, audit and cost attribution to the call. See LLM cost attribution and budgets.
In DVARA, if you haven't set your own price for a model, its calls are costed at the price in the catalogue that ships with your DVARA version. A price you set in Flightdeck always wins. You can switch catalogue prices off. Then a model with no price you set is counted as costing nothing. Every cost record says which price it used.
What should I check before relying on this data?
The weekly review marked this entry. A person saw each mark and kept the model listed.
- The provider charges more above 200K tokens of input; the table shows the base price.
Which other Google Gemini models are there?
- Deep Research Preview (Apr-21-2026)$2 in · $12 out · 131K context
- Gemini 2.5 Computer Use Preview 10-2025$1.25 in · $10 out · 128K context
- Gemini 2.5 Flash$0.3 in · $2.5 out · 1.05M context
- Nano Banana$0.3 in · $30 out · 33K context
- Gemini 2.5 Flash-Lite$0.1 in · $0.4 out · 1.05M context
- Gemini 2.5 Flash Preview TTS$0.5 in · $10 out · 8K context
- Gemini 2.5 Pro$1.25 in · $10 out · 1.05M context
- Gemini 2.5 Pro Preview TTS$1 in · $20 out · 8K context
- Gemini 3 Flash Preview$0.5 in · $3 out · 1.05M context
- Nano Banana Pro$2 in · $120 out · 66K context
- Nano Banana Pro$2 in · $120 out · 131K context
- Nano Banana 2$0.5 in · $60 out · 131K context
Which models from other providers are similar to Deep Research Max Preview (Apr-21-2026)?
These are the closest in context window and input price, one per provider.
- Mistral Large 2.1Mistral AI · $2 in · $6 out · 131K context
- Pixtral Large (25.02) (EU)Amazon Bedrock · $2 in · $6 out · 128K context
- Qwen2.5-VL 72B InstructAlibaba Qwen · $2.294 in · $6.881 out · 131K context
- GPT-5.2 ChatOpenAI · $1.75 in · $14 out · 128K context
- Command A PlusCohere · $2.5 in · $10 out · 128K context