- Third-party
- Zero data retention
Gemini 3 Flash is Google's fast multimodal model with frontier intelligence, superior search, and grounding capabilities.
| Model Info | |
|---|---|
| Context Window ↗ | 1,000,000 tokens |
| Terms and License | link ↗ |
| More information | link ↗ |
| Zero data retention | Yes |
| Request formats | Chat Completions |
| Pricing |
|
const response = await env.AI.run(
'google/gemini-3-flash',
{ contents: [{ parts: [{ text: 'What are the three laws of thermodynamics?' }], role: 'user' }] },
)
console.log(response)curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"model": "google/gemini-3-flash",
"input": {
"contents": [
{
"parts": [
{
"text": "What are the three laws of thermodynamics?"
}
],
"role": "user"
}
]
}
}'While there are technically four laws (starting with the "Zeroth Law"), the **Three Laws of Thermodynamics** are the fundamental principles that describe how energy, heat, and matter behave.
Here is a breakdown of the three laws:
---
### 1. The First Law: Conservation of Energy
**Definition:** Energy cannot be created or destroyed, only transformed from one form to another. The total energy of an isolated system remains constant.
* **In simple terms:** You can't get something for nothing.
* **The Equation:** $\Delta U = Q - W$
* ($\Delta U$ is the change in internal energy, $Q$ is heat added, and $W$ is work done by the system).
* **Example:** In a car engine, the chemical energy of the gasoline is converted into heat and then into mechanical work to move the car.
### 2. The Second Law: Entropy
**Definition:** The total entropy (disorder) of an isolated system can never decrease over time; it can only remain constant or increase. This law dictates the "arrow of time."
* **In simple terms:** Things tend to get messy and disorganized over time. Heat always flows spontaneously from a hotter object to a colder object, never the other way around without adding work.
* **Key Concept:** This law explains why 100% efficiency is impossible—some energy is always "lost" as waste heat.
* **Example:** If you place a hot cup of coffee in a cold room, the coffee cools down as heat spreads into the room. The heat will never spontaneously jump back from the room into the coffee to make it boil again.
### 3. The Third Law: Absolute Zero
**Definition:** As the temperature of a system approaches absolute zero ($0$ Kelvin or $-273.15^\circ$ Celsius), the entropy of a perfect crystal approaches a constant minimum (usually zero).
* **In simple terms:** You can’t reach absolute zero.
* **Key Concept:** As atoms get colder, they slow down. At absolute zero, all molecular motion would essentially stop. However, because moving heat requires a temperature difference, it is physically impossible to reach a temperature of exactly zero Kelvin in a finite number of steps.
* **Example:** Scientists use advanced lasers and magnets to get atoms within billionths of a degree of absolute zero, but they can never quite hit the "stop" button entirely.
---
### Summary Mnemonic
There is a famous (and slightly cynical) way to remember these laws, often called **"Ginsberg’s Theorem"**:
1. **First Law:** You can't win. (You can't get more energy out than you put in).
2. **Second Law:** You can't break even. (You always lose some energy to entropy).
3. **Third Law:** You can't get out of the game. (You can't reach absolute zero to stop the process).
***
**Note on the Zeroth Law:** If you ever hear about a "Zeroth Law," it simply states that if two systems are each in thermal equilibrium with a third system, they are in thermal equilibrium with each other. This is the law that allows us to use thermometers.{
"candidates": [
{
"content": {
"role": "model",
"parts": [
{
"text": "While there are technically four laws (starting with the \"Zeroth Law\"), the **Three Laws of Thermodynamics** are the fundamental principles that describe how energy, heat, and matter behave.\n\nHere is a breakdown of the three laws:\n\n---\n\n### 1. The First Law: Conservation of Energy\n**Definition:** Energy cannot be created or destroyed, only transformed from one form to another. The total energy of an isolated system remains constant.\n\n* **In simple terms:** You can't get something for nothing.\n* **The Equation:** $\\Delta U = Q - W$\n * ($\\Delta U$ is the change in internal energy, $Q$ is heat added, and $W$ is work done by the system).\n* **Example:** In a car engine, the chemical energy of the gasoline is converted into heat and then into mechanical work to move the car.\n\n### 2. The Second Law: Entropy\n**Definition:** The total entropy (disorder) of an isolated system can never decrease over time; it can only remain constant or increase. This law dictates the \"arrow of time.\"\n\n* **In simple terms:** Things tend to get messy and disorganized over time. Heat always flows spontaneously from a hotter object to a colder object, never the other way around without adding work.\n* **Key Concept:** This law explains why 100% efficiency is impossible—some energy is always \"lost\" as waste heat.\n* **Example:** If you place a hot cup of coffee in a cold room, the coffee cools down as heat spreads into the room. The heat will never spontaneously jump back from the room into the coffee to make it boil again.\n\n### 3. The Third Law: Absolute Zero\n**Definition:** As the temperature of a system approaches absolute zero ($0$ Kelvin or $-273.15^\\circ$ Celsius), the entropy of a perfect crystal approaches a constant minimum (usually zero).\n\n* **In simple terms:** You can’t reach absolute zero.\n* **Key Concept:** As atoms get colder, they slow down. At absolute zero, all molecular motion would essentially stop. However, because moving heat requires a temperature difference, it is physically impossible to reach a temperature of exactly zero Kelvin in a finite number of steps.\n* **Example:** Scientists use advanced lasers and magnets to get atoms within billionths of a degree of absolute zero, but they can never quite hit the \"stop\" button entirely.\n\n---\n\n### Summary Mnemonic\nThere is a famous (and slightly cynical) way to remember these laws, often called **\"Ginsberg’s Theorem\"**:\n\n1. **First Law:** You can't win. (You can't get more energy out than you put in).\n2. **Second Law:** You can't break even. (You always lose some energy to entropy).\n3. **Third Law:** You can't get out of the game. (You can't reach absolute zero to stop the process).\n\n***\n\n**Note on the Zeroth Law:** If you ever hear about a \"Zeroth Law,\" it simply states that if two systems are each in thermal equilibrium with a third system, they are in thermal equilibrium with each other. This is the law that allows us to use thermometers.",
"thoughtSignature": "AY89a1+RdEkQqSCfe6KTvKWMknfIBywa5lOKfzJSDw4DHkMrF7/+jEpXKC9RABDVwVPQKy7Q9lCrPUvMBcuV858jRzDwHe88wDTSzoCgfjF7ztathjHzYzjkUyPg61Eg46vOVPb2f5Ca/bCUkdl4tvbk04gv/GueV57KO0wJbnF2oBjX/estVrz2k9mQCX/oSEEQDUIYg4BwjzqAlK7vQaIePTbsapXiEBugPRd2/rfscLbiEnwvx66ZmpCsxRpnzWIuw8biuQBTmqU5tzA8eJ7wez0w1O6KX+kcrnA9wEAqT0MF+S2qhF3Sc28320v9a6JoFbSdzUL7MccJ75M1ruo25NgnBFzckrPCBtjFGJXKssdBtmCKCfZrKFa7s8y72DjpJyUqcg4rilk+WiBMFmUulrUpFXu+F2VedL0tSvAOx7ih0y5DWoNmzClQBuF/aoUROy1ey7Kq0HD0eySV2yK0dy2TN28XITvGhhq7I6w+TUMNiOjprZN5mGx97l/WpZ0KI+lDjqabo7QAJY7a/Oe2hi6Fycfq/vNj5ntB2geJZqqtrI2MZF3g6mJg4D4pf1nfUMPJvESuubqHTbHEeIgXqIpE/k/z6R6ffQNg/8ACLBQHP+dGcC8gIx3rMOt6e02RvzlJAOh0rFTKiAckjCZ3SOF/8QLXdaEQ2kmaXw6j+uMp5WWJpDFu7ReWlhtwQbCFIcgrkwSfXPjwl50tUAYJQ6UnVIPItrd6bNd260QRUJTqJjFV7aISbqRWbZL1qXGGGKSuSkRv5C2srVxY7emv03AbH+NFYCenLsPgLkf2zFOT3+H9rl+MAO/gtnAc8IELv+cH89fIDjhBOWNIPItNCVlnIAmMufQY5NmdzqhWrt6h2ZsOOIEwjTci4m3K/bBuuHrLp9fybz4ubaiWBJCsFIBrVzt5j6GJoduMHJ+98uJ0woNl0EkcmClV7mKa4KR7WrN0lU6EB/lCkleHMRn0zWvXIqNetPqYijeYB8rsCyRi/ujqFZAWmveLRx2UBbgvtcNRruxGY+mg7aGXGsGbPdQLOGMPbjhpH0YXTeS6rMPglF4zOq2nvkISug5NxUIOK6OiupZYar/6aDrXODsPBr45HhrYz4JMJA+9JzzYHgueMMu5Iw+KL+bPGgXXoBvO/XrNMnO9rTevi6eVHD/0jrJHCu6fG87lTCBDYjF0JQYekrqkTf4WbBfEpyVPKm2SP4lgLdyLN/E2rieXDi3srBr6NP1C6JDyQlDXuWPVd27npP1Zbtfm2nUnq73OmMJKIzQEmRSSok0SFg2YvOw1gAe4D/smJTjU+1pwQXQlqQxwkjJ1DuC2IYwPR9hIfxex0aZqY0gXmhDotHIClRD8uPCpW82kLcy69evbvjRdMy6bDG3ytIhOiQNeBy5VMJ1tkPtGXcAgZPcGnV56+tTwbuChI75Vonk8kGZRK7c79+CbP3Hm+FYe2R53k+uPGOJgRiCYNjRzuS66R7u3NvzNnz/F1FKPTW/cosrusLx3ghPcfvhawu7O9xUDhP6uyUgN3n62gluc7TXHbRRiUQFF33A7noDGZC8X1M2BpvYivm/IMeFlBDsJDiGQgVNK5fYCWz+rjfxC7Q6eFFcrs9TEwvKnsrTtxjuAIVC5eng6sH/bjYpexZ0egORleQOQ+MpGqVx7lhK9ILLtXNkbhIpyvc7RgDyckCt8ei9HvG5kp1HZ6GNn3AE+MoCpVO+0HTNCV8f3SKuXI+SEHR1/2DyXq1h8MDCYLrrEDiyhylPsZ5wOQqrNRtxQQ8hBTG8TB8/dK9WbPn9d9uzEMxJ0FuCVW+phxy30wfrfPgK1VF+h44LG4ouCDXXn3ulIBpdVPGo2oZJFOuLtcJSreXyHklkQX9nVT6HmUtRafT82+78DWAm5chkZ933SJsSabovTAj2uVuMXgDD30Gpk7oJ6wuSKw8vjbPk/0i6Ep8L21Mv++Kuj3WskXQlEtXgujOiQx/CdpTmlTbjeQ6hv0/bagQVuEN1RtNRAS6lmyCqcuo2VuaUqwDvGjRT6+/OM/7memyezkWmJAufx7vgvV+BmlIMUCcNQKRFT"
}
]
},
"finishReason": "STOP"
}
],
"usageMetadata": {
"promptTokenCount": 8,
"candidatesTokenCount": 697,
"totalTokenCount": 1107,
"trafficType": "ON_DEMAND",
"promptTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 8
}
],
"candidatesTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 697
}
],
"thoughtsTokenCount": 402
},
"modelVersion": "gemini-3-flash-preview",
"createTime": "2026-07-24T21:49:35.132185Z",
"responseId": "791jatmICOSg3dAPo6eioAg",
"gatewayMetadata": {
"keySource": "Unified"
}
}With System Instruction — Using a system instruction to set context
const response = await env.AI.run(
'google/gemini-3-flash',
{
contents: [{ parts: [{ text: 'How do I read a JSON file in Python?' }], role: 'user' }],
generationConfig: { temperature: 0.3 },
systemInstruction: {
parts: [{ text: 'You are a helpful coding assistant specializing in Python.' }],
},
},
)
console.log(response)curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"model": "google/gemini-3-flash",
"input": {
"contents": [
{
"parts": [
{
"text": "How do I read a JSON file in Python?"
}
],
"role": "user"
}
],
"generationConfig": {
"temperature": 0.3
},
"systemInstruction": {
"parts": [
{
"text": "You are a helpful coding assistant specializing in Python."
}
]
}
}
}'To read a JSON file in Python, you use the built-in `json` module.
Here is the most common and recommended way to do it:
### The Standard Method
```python
import json
# Open the file and load the data
with open('data.json', 'r', encoding='utf-8') as file:
data = json.load(file)
# Now 'data' is a Python dictionary or list
print(data)
```
### Breakdown of the code:
1. **`import json`**: Imports Python’s built-in JSON library.
2. **`with open(...)`**: This is a "context manager." It ensures the file is properly closed after you are done reading it, even if an error occurs.
3. **`'r'`**: Opens the file in **read** mode.
4. **`encoding='utf-8'`**: This is optional but highly recommended to ensure special characters (like emojis or non-English letters) are handled correctly.
5. **`json.load(file)`**: This function takes the file object and converts the JSON data into a Python object (usually a `dict` or a `list`).
---
### `json.load()` vs `json.loads()`
It is easy to confuse these two functions:
* **`json.load()`** (no 's'): Used to read from a **file object**.
* **`json.loads()`** (stands for "load string"): Used to parse a **string** containing JSON data.
**Example of `json.loads()`:**
```python
import json
json_string = '{"name": "Alice", "age": 30}'
data = json.loads(json_string)
print(data["name"]) # Output: Alice
```
---
### Handling Errors
If the file doesn't exist or the JSON is formatted incorrectly, your program will crash. You can wrap the code in a `try-except` block to handle this gracefully:
```python
import json
try:
with open('data.json', 'r') as file:
data = json.load(file)
except FileNotFoundError:
print("The file was not found.")
except json.JSONDecodeError:
print("The file is not a valid JSON format.")
```
### Pro Tip: Using Pandas (for Data Science)
If your JSON file contains a list of records (like a table), the **Pandas** library is often much faster and easier:
```python
import pandas as pd
df = pd.read_json('data.json')
print(df.head())
```{
"candidates": [
{
"content": {
"role": "model",
"parts": [
{
"text": "To read a JSON file in Python, you use the built-in `json` module.\n\nHere is the most common and recommended way to do it:\n\n### The Standard Method\n```python\nimport json\n\n# Open the file and load the data\nwith open('data.json', 'r', encoding='utf-8') as file:\n data = json.load(file)\n\n# Now 'data' is a Python dictionary or list\nprint(data)\n```\n\n### Breakdown of the code:\n1. **`import json`**: Imports Python’s built-in JSON library.\n2. **`with open(...)`**: This is a \"context manager.\" It ensures the file is properly closed after you are done reading it, even if an error occurs.\n3. **`'r'`**: Opens the file in **read** mode.\n4. **`encoding='utf-8'`**: This is optional but highly recommended to ensure special characters (like emojis or non-English letters) are handled correctly.\n5. **`json.load(file)`**: This function takes the file object and converts the JSON data into a Python object (usually a `dict` or a `list`).\n\n---\n\n### `json.load()` vs `json.loads()`\nIt is easy to confuse these two functions:\n* **`json.load()`** (no 's'): Used to read from a **file object**.\n* **`json.loads()`** (stands for \"load string\"): Used to parse a **string** containing JSON data.\n\n**Example of `json.loads()`:**\n```python\nimport json\n\njson_string = '{\"name\": \"Alice\", \"age\": 30}'\ndata = json.loads(json_string)\n\nprint(data[\"name\"]) # Output: Alice\n```\n\n---\n\n### Handling Errors\nIf the file doesn't exist or the JSON is formatted incorrectly, your program will crash. You can wrap the code in a `try-except` block to handle this gracefully:\n\n```python\nimport json\n\ntry:\n with open('data.json', 'r') as file:\n data = json.load(file)\nexcept FileNotFoundError:\n print(\"The file was not found.\")\nexcept json.JSONDecodeError:\n print(\"The file is not a valid JSON format.\")\n```\n\n### Pro Tip: Using Pandas (for Data Science)\nIf your JSON file contains a list of records (like a table), the **Pandas** library is often much faster and easier:\n\n```python\nimport pandas as pd\n\ndf = pd.read_json('data.json')\nprint(df.head())\n```",
"thoughtSignature": "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"
}
]
},
"finishReason": "STOP"
}
],
"usageMetadata": {
"promptTokenCount": 20,
"candidatesTokenCount": 582,
"totalTokenCount": 875,
"trafficType": "ON_DEMAND",
"promptTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 20
}
],
"candidatesTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 582
}
],
"thoughtsTokenCount": 273
},
"modelVersion": "gemini-3-flash-preview",
"createTime": "2026-07-24T21:49:42.962953Z",
"responseId": "9t1jaonjOoGL3dAPnayyuA0",
"gatewayMetadata": {
"keySource": "Unified"
}
}Multi-turn Conversation — Continuing a conversation with context
const response = await env.AI.run(
'google/gemini-3-flash',
{
contents: [
{
parts: [{ text: 'I need help planning a road trip from San Francisco to Los Angeles.' }],
role: 'user',
},
{
parts: [
{
text: "I'd be happy to help! The drive is about 380 miles and takes roughly 5-6 hours. Would you like suggestions for scenic routes or interesting stops along the way?",
},
],
role: 'model',
},
{ parts: [{ text: 'Yes, what are some good places to stop?' }], role: 'user' },
],
generationConfig: { maxOutputTokens: 2048 },
},
)
console.log(response)curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"model": "google/gemini-3-flash",
"input": {
"contents": [
{
"parts": [
{
"text": "I need help planning a road trip from San Francisco to Los Angeles."
}
],
"role": "user"
},
{
"parts": [
{
"text": "I'\''d be happy to help! The drive is about 380 miles and takes roughly 5-6 hours. Would you like suggestions for scenic routes or interesting stops along the way?"
}
],
"role": "model"
},
{
"parts": [
{
"text": "Yes, what are some good places to stop?"
}
],
"role": "user"
}
],
"generationConfig": {
"maxOutputTokens": 2048
}
}
}'The stops you choose will depend entirely on which route you take. Since I mentioned the 5-6 hour timeframe earlier, that usually refers to the **I-5 (Inland)** route, but most people prefer the **Highway 1 (Coastal)** route for a "road trip" experience. Here are the best stops for the three most common routes: ### 1. The Scenic Route: Highway 1 (Pacific Coast Highway) *Best for: Incredible views, photo ops, and a multi-day trip.* * **Santa Cruz:** Walk the historic Beach Boardwalk or see the Redwood trees at Roaring Camp. * **Monterey & Carmel-by-the-Sea:** Visit the world-famous Monterey Bay Aquarium and drive the "17-Mile Drive" for views of Pebble Beach. * **Big Sur:** Cross the iconic **Bixby Creek Bridge** and stop at **McWay Falls** (a waterfall that drops right onto the beach). * **San Simeon:** Tour **Hearst Castle**, the massive estate of William Randolph Hearst, or see the Elephant Seal Rookery at Piedras Blancas. * **Solvang:** A charming, "Little Denmark" village known for Dutch architecture and bakeries. * **Santa Barbara:** Known as the "American Riviera," it has beautiful Spanish architecture and great shopping on State Street. ### 2. The Middle Ground: US-101 *Best for: A balance of speed and scenery (approx. 6–7 hours).* * **Paso Robles:** If you like wine, this is one of California’s premier wine regions with beautiful rolling hills. * **San Luis Obispo (SLO):** Visit the historic mission, walk through the "Bubblegum Alley," or hike Bishop Peak. * **Pismo Beach:** Famous for its wide sandy beaches and "Clam Chowder" at Splash Café. * **Ventura:** A great spot to see the pier and grab fresh seafood before hitting the LA traffic. ### 3. The Quickest Route: I-5 (The Grapevine) *Best for: Getting there as fast as possible (approx. 5–6 hours).* * **Harris Ranch (Coalinga):** A famous landmark and massive ranch. It's the best place for a high-quality steak dinner or a deli sandwich in the middle of the valley. * **Bravo Farms (Kettleman City):** A fun, kitschy stop with a petting zoo, ice cream, nuts, and a "seven-story" treehouse for kids to climb. * **Tejon Ranch Outlets:** Located at the base of "The Grapevine," it’s great for discount designer shopping before the final stretch into LA. --- ### A few "Pro Tips" for your drive: 1. **Check for Closures:** If you take **Highway 1 (Big Sur)**, check Caltrans before you go. That road is prone to landslides and parts are occasionally closed for repairs. 2. **The "Grapevine":** If taking the I-5, you will pass over a high mountain ridge called the Grapevine. In winter, it can occasionally close due to snow; in summer, keep an eye on your engine temperature as it's a long, steep climb. 3. **LA Traffic:** Try to time your arrival in Los Angeles either before 2:00 PM or after 8:00 PM to avoid the worst of the gridlock. **Which route sounds more like your style?** I can give you more specific recommendations for food or hotels once you choose!
{
"candidates": [
{
"content": {
"role": "model",
"parts": [
{
"text": "The stops you choose will depend entirely on which route you take. Since I mentioned the 5-6 hour timeframe earlier, that usually refers to the **I-5 (Inland)** route, but most people prefer the **Highway 1 (Coastal)** route for a \"road trip\" experience.\n\nHere are the best stops for the three most common routes:\n\n### 1. The Scenic Route: Highway 1 (Pacific Coast Highway)\n*Best for: Incredible views, photo ops, and a multi-day trip.*\n* **Santa Cruz:** Walk the historic Beach Boardwalk or see the Redwood trees at Roaring Camp.\n* **Monterey & Carmel-by-the-Sea:** Visit the world-famous Monterey Bay Aquarium and drive the \"17-Mile Drive\" for views of Pebble Beach.\n* **Big Sur:** Cross the iconic **Bixby Creek Bridge** and stop at **McWay Falls** (a waterfall that drops right onto the beach).\n* **San Simeon:** Tour **Hearst Castle**, the massive estate of William Randolph Hearst, or see the Elephant Seal Rookery at Piedras Blancas.\n* **Solvang:** A charming, \"Little Denmark\" village known for Dutch architecture and bakeries.\n* **Santa Barbara:** Known as the \"American Riviera,\" it has beautiful Spanish architecture and great shopping on State Street.\n\n### 2. The Middle Ground: US-101\n*Best for: A balance of speed and scenery (approx. 6–7 hours).*\n* **Paso Robles:** If you like wine, this is one of California’s premier wine regions with beautiful rolling hills.\n* **San Luis Obispo (SLO):** Visit the historic mission, walk through the \"Bubblegum Alley,\" or hike Bishop Peak.\n* **Pismo Beach:** Famous for its wide sandy beaches and \"Clam Chowder\" at Splash Café.\n* **Ventura:** A great spot to see the pier and grab fresh seafood before hitting the LA traffic.\n\n### 3. The Quickest Route: I-5 (The Grapevine)\n*Best for: Getting there as fast as possible (approx. 5–6 hours).*\n* **Harris Ranch (Coalinga):** A famous landmark and massive ranch. It's the best place for a high-quality steak dinner or a deli sandwich in the middle of the valley.\n* **Bravo Farms (Kettleman City):** A fun, kitschy stop with a petting zoo, ice cream, nuts, and a \"seven-story\" treehouse for kids to climb.\n* **Tejon Ranch Outlets:** Located at the base of \"The Grapevine,\" it’s great for discount designer shopping before the final stretch into LA.\n\n---\n\n### A few \"Pro Tips\" for your drive:\n1. **Check for Closures:** If you take **Highway 1 (Big Sur)**, check Caltrans before you go. That road is prone to landslides and parts are occasionally closed for repairs.\n2. **The \"Grapevine\":** If taking the I-5, you will pass over a high mountain ridge called the Grapevine. In winter, it can occasionally close due to snow; in summer, keep an eye on your engine temperature as it's a long, steep climb.\n3. **LA Traffic:** Try to time your arrival in Los Angeles either before 2:00 PM or after 8:00 PM to avoid the worst of the gridlock.\n\n**Which route sounds more like your style?** I can give you more specific recommendations for food or hotels once you choose!",
"thoughtSignature": "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"
}
]
},
"finishReason": "STOP"
}
],
"usageMetadata": {
"promptTokenCount": 64,
"candidatesTokenCount": 762,
"totalTokenCount": 1356,
"trafficType": "ON_DEMAND",
"promptTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 64
}
],
"candidatesTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 762
}
],
"thoughtsTokenCount": 530
},
"modelVersion": "gemini-3-flash-preview",
"createTime": "2026-07-24T21:49:48.502515Z",
"responseId": "_N1javPVHrSk3dAP45z02AQ",
"gatewayMetadata": {
"keySource": "Unified"
}
}Creative Writing — Higher temperature for creative output
const response = await env.AI.run(
'google/gemini-3-flash',
{
contents: [
{
parts: [{ text: 'Write a short story opening about a detective finding an unusual clue.' }],
role: 'user',
},
],
generationConfig: { maxOutputTokens: 4096, temperature: 0.8 },
},
)
console.log(response)curl https://api.cloudflare.com/client/v4/accounts/$CLOUDFLARE_ACCOUNT_ID/ai/run \
--header "Authorization: Bearer $CLOUDFLARE_API_TOKEN" \
--header "Content-Type: application/json" \
--data '{
"model": "google/gemini-3-flash",
"input": {
"contents": [
{
"parts": [
{
"text": "Write a short story opening about a detective finding an unusual clue."
}
],
"role": "user"
}
],
"generationConfig": {
"maxOutputTokens": 4096,
"temperature": 0.8
}
}
}'The humidity in the room was wrong. That was the first thing Detective Elias Thorne noticed. It wasn’t the metallic tang of blood or the heavy stillness of a man who had stopped breathing three hours ago. It was the smell of a thunderstorm in a room with no open windows. Julian Vane, a man who had made millions selling security software, sat slumped in his designer leather chair. His throat had been opened with surgical precision, but Thorne’s eyes didn't linger on the gore. They were drawn to the mahogany desk, where a small, silver saucer sat perfectly centered. In the middle of the saucer was a single, glowing sugar cube. Thorne leaned in, his shadow stretching long across the Persian rug. He didn’t reach for his tweezers yet. The cube wasn't just white; it pulsed with a soft, rhythmic violet light, timed exactly to the pace of a resting human heartbeat. "Miller," Thorne called out without looking back at his partner. "Tell me the victim had a penchant for bioluminescent snacks." "He was a hedge fund manager, Elias," Miller replied, snapping photos of the entryway. "The most exotic thing he consumed was unpasteurized goat cheese." Thorne pulled a magnifying loupe from his coat pocket. As he peered through the glass, his breath hitched. Inside the crystalline structure of the sugar, frozen like a prehistoric insect in amber, was a microscopic set of brass gears. They were no larger than a grain of dust, and they were turning slowly in reverse. It wasn't just a clue. It was a piece of machinery that shouldn't exist, ticking away inside a condiment. "Get the lead-lined evidence bag," Thorne whispered, the hair on his arms standing up as the violet pulse of the cube suddenly skipped a beat, matching his own racing heart. "I don't think we're looking for a murderer anymore. I think we're looking for a clockmaker."
{
"candidates": [
{
"content": {
"role": "model",
"parts": [
{
"text": "The humidity in the room was wrong. That was the first thing Detective Elias Thorne noticed. It wasn’t the metallic tang of blood or the heavy stillness of a man who had stopped breathing three hours ago. It was the smell of a thunderstorm in a room with no open windows.\n\nJulian Vane, a man who had made millions selling security software, sat slumped in his designer leather chair. His throat had been opened with surgical precision, but Thorne’s eyes didn't linger on the gore. They were drawn to the mahogany desk, where a small, silver saucer sat perfectly centered.\n\nIn the middle of the saucer was a single, glowing sugar cube.\n\nThorne leaned in, his shadow stretching long across the Persian rug. He didn’t reach for his tweezers yet. The cube wasn't just white; it pulsed with a soft, rhythmic violet light, timed exactly to the pace of a resting human heartbeat. \n\n\"Miller,\" Thorne called out without looking back at his partner. \"Tell me the victim had a penchant for bioluminescent snacks.\"\n\n\"He was a hedge fund manager, Elias,\" Miller replied, snapping photos of the entryway. \"The most exotic thing he consumed was unpasteurized goat cheese.\"\n\nThorne pulled a magnifying loupe from his coat pocket. As he peered through the glass, his breath hitched. Inside the crystalline structure of the sugar, frozen like a prehistoric insect in amber, was a microscopic set of brass gears. They were no larger than a grain of dust, and they were turning slowly in reverse.\n\nIt wasn't just a clue. It was a piece of machinery that shouldn't exist, ticking away inside a condiment. \n\n\"Get the lead-lined evidence bag,\" Thorne whispered, the hair on his arms standing up as the violet pulse of the cube suddenly skipped a beat, matching his own racing heart. \"I don't think we're looking for a murderer anymore. I think we're looking for a clockmaker.\"",
"thoughtSignature": "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"
}
]
},
"finishReason": "STOP"
}
],
"usageMetadata": {
"promptTokenCount": 13,
"candidatesTokenCount": 409,
"totalTokenCount": 1275,
"trafficType": "ON_DEMAND",
"promptTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 13
}
],
"candidatesTokensDetails": [
{
"modality": "TEXT",
"tokenCount": 409
}
],
"thoughtsTokenCount": 853
},
"modelVersion": "gemini-3-flash-preview",
"createTime": "2026-07-24T23:10:39.662413Z",
"responseId": "7_Bjao23KPaH3dAP_vWlwAY",
"gatewayMetadata": {
"keySource": "Unified"
}
}▶contents[]
arrayrequired▶systemInstruction{}
object▶generationConfig{}
object▶safetySettings[]
array▶tools[]
arraytoolConfig
▶messages[]
arrayrequiredtemperature
numberminimum: 0maximum: 2max_tokens
numberexclusiveMinimum: 0max_completion_tokens
numberexclusiveMinimum: 0top_p
numberminimum: 0maximum: 1frequency_penalty
numberminimum: -2maximum: 2presence_penalty
numberminimum: -2maximum: 2stream
boolean▶stream_options{}
object▶tools[]
arraytool_choice
response_format
▶modalities[]
array▶audio{}
objectreasoning_effort
stringOptional reasoning control; availability and accepted values are model-dependent.▶candidates[]
array▶usageMetadata{}
objectmodelVersion
stringid
stringobject
stringcreated
numbermodel
string▶choices[]
array▶usage{}
object