browse · models · Google
Google model — context 2M, capabilities: vision · tool_use · thinking.
| Model ID | gemini-3-pro |
| Provider | |
| Family | gemini-3 |
| Status | active |
| Input $/MTok | $2.5 |
| Output $/MTok | $10 |
| Cache read $/MTok | — |
| Cache write $/MTok | — |
| Context window | 2,000,000 tokens (2M) |
| Max output | 64,000 tokens (64K) |
| Input modalities | textimage |
| Output modalities | text |
| Capabilities | visiontool_usethinking |
| Supported parameters | max_tokenstemperaturetop_ptoolstool_choicereasoninginclude_reasoning |
| Knowledge cutoff | 2025-09 |
| Release date | 2025-09-01 |
| Last synced | 2026-05-13 |
| Deprecated | — |
| Official docs | Open docs |
| Intelligence index | not measured |
| Throughput | — |
// claude-desktop / cursor / windsurf
{
"jsonrpc": "2.0",
"method": "tools/call",
"id": 1,
"params": {
"name": "get_model_pricing",
"arguments": { "model_id": "gemini-3-pro" }
}
}curl -X POST https://mcp.aiworkflowmcp.com/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","method":"tools/call","id":1,"params":{"name":"get_model_pricing","arguments":{"model_id":"gemini-3-pro"}}}'