Gemini 1.5 Pro API Rate Limits, Pricing & Performance (July 2026)

Rate limits by tier

Gemini 1.5 Pro is shut down. The limits below describe its current successor, Gemini 2.5 Pro (api_name: gemini-2.5-pro). Google does not publish stable RPM/TPM numbers — active quotas are project-specific and shown in AI Studio console.

Swipe the table sideways to see every limit column.

• 1.Google no longer publishes stable RPM/TPM numbers for Gemini models. Active limits are project-specific and visible in AI Studio dashboard.
• 2.The 200K long-context surcharge applies to the WHOLE prompt — crossing the threshold by even one token roughly doubles input cost and increases output cost by 50%.
• 3.Thinking tokens (Gemini 2.5+) are billed at the output rate. Disable thinking for tasks that do not require step-by-step reasoning.
• 4.Original Gemini 1.5 Pro pricing is not published (model is retired). Prices above are for successor Gemini 2.5 Pro.

Limits verified against the Google docs, July 10, 2026 — superseded model; data below describes successor Gemini 2.5 Pro.

Token pricing

What you pay per million tokens (USD). Input and output are billed separately.

• Pricing is for successor Gemini 2.5 Pro (≤200K token requests). Gemini 1.5 Pro pricing is not published (retired).
• Long-context surcharge: input rises to $2.50/MTok and output to $15.00/MTok for prompts exceeding 200K tokens. Surcharge applies to the ENTIRE prompt.
• Context cache reads: ~$0.125/MTok (~10% of standard input price). Cache storage billed separately.
• Prices raised July 2, 2026. Verify current rates at ai.google.dev/pricing.
• Thinking tokens billed at output rate ($10.00/MTok for ≤200K). Set thinking budgets to control output-token volume.

Run your own numbers

Drag your real monthly token volumes and watch the bill update live — priced against rival models at the same usage.

gemini-1.5-pro (retired; use gemini-2.5-pro) vs the alternatives

Gemini 2.5 Pro (successor) vs leading multimodal APIs on context, price, and capabilities.

Swipe the table sideways to see every model.

Hitting a 429? The playbook

1// retry.ts — Gemini 2.5 Pro (successor to 1.5 Pro) with exponential backoff
2// IMPORTANT: use 'gemini-2.5-pro' — 'gemini-1.5-pro' is retired and will return errors
3
4const API_KEY = process.env.GOOGLE_API_KEY!;
5const MODEL = 'gemini-2.5-pro'; // NOT gemini-1.5-pro (retired)
6const BASE_URL = `https://generativelanguage.googleapis.com/v1beta/models/${MODEL}:generateContent`;
7
8async function generateContent(prompt: string, maxRetries = 5): Promise<unknown> {
9  let delay = 1000;
10  for (let attempt = 0; attempt <= maxRetries; attempt++) {
11    const response = await fetch(`${BASE_URL}?key=${API_KEY}`, {
12      method: 'POST',
13      headers: { 'Content-Type': 'application/json' },
14      body: JSON.stringify({
15        contents: [{ role: 'user', parts: [{ text: prompt }] }],
16      }),
17    });
18    if (response.ok) return response.json();
19    if (response.status === 404) {
20      throw new Error('Model not found — ensure you are using gemini-2.5-pro, not gemini-1.5-pro (retired)');
21    }
22    if (response.status !== 429 || attempt === maxRetries) {
23      throw new Error(`Gemini 2.5 Pro error ${response.status}: ${await response.text()}`);
24    }
25    const retryAfter = response.headers.get('Retry-After');
26    const jitter = 0.8 + Math.random() * 0.4;
27    const waitMs = retryAfter ? parseInt(retryAfter) * 1000 : delay * jitter;
28    console.warn(`Rate limited. Retrying in ${Math.round(waitMs)}ms (attempt ${attempt + 1})`);
29    await new Promise((r) => setTimeout(r, waitMs));
30    delay = Math.min(delay * 2, 64_000);
31  }
32  throw new Error('Max retries exceeded');
33}

How to raise your limits

The ladder from the starter tier to enterprise — what each rung takes, and what it unlocks.

Cut your token spend

Related models

More gemini-1.5-pro (retired; use gemini-2.5-pro) resources