# Gemini 3.6 Flash

Gemini 3.6 Flash is an AI model released by Google on Jul 21 2026. It has a 1M context window. Tracked results include 63.9% on MLE-Bench, 39% on BullshitBench v2 and 60.4% on ARC-AGI-2.

## Facts

| Field | Value |
| --- | --- |
| Model | Gemini 3.6 Flash |
| Developer | Google |
| Release date | Tuesday, Jul 21 2026 |
| Licensing | Closed |
| Context window | 1M |

## API pricing

All rates in USD per 1,000,000 tokens, pay-as-you-go.

| Tier | Input | Cached input | Output |
| --- | --- | --- | --- |
| Standard, through 2026-12-31 | $0.75 | $0.075 | $3.75 |

Output prices include reasoning tokens.
Verified October 19, 2026 against the first-party source: https://ai.google.dev/gemini-api/docs/pricing

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| BullshitBench v2 | 39% | [BullshitBench](https://github.com/petergpt/bullshit-benchmark) | Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. |
| Gray Swan IPI (k = 1) | 7.3% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better. |
| Gray Swan IPI (k = 10) | 32.2% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. |
| Gray Swan IPI (k = 15) | 37.3% | Lab | Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. |
| ProgramBench | 0.5% | [ProgramBench](https://programbench.com/), retrieved 2026-09-11 | The AI receives a working program and its documentation, then builds a replacement from scratch without the original source code, internet access or decompilation. The score is the percentage of 200 programs that pass every behavioral test. We record each model's best published mini-SWE-agent result, including higher reasoning efforts where available. Partial test-pass rates and almost-solved programs do not count toward this score. Equal scores share a rank here; the official board also uses partial progress to break ties. Higher is better. |
| DeepSWE 1.1 | 49% | Lab | Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. |
| MLE-Bench | 63.9% | Lab | Can the AI do the work of a machine-learning engineer? It competes in real Kaggle competitions — building, training, and tuning models end to end — and the score reflects how well it places. Higher is better. |
| BU Bench | 68% | Lab | Can the AI drive a real web browser to finish tasks — clicking, filling forms, and navigating sites the way a person would? Run by Browser Use on their BU Bench task set. Higher is better. |
| ARC-AGI-2 | 60.4% | [BenchLM](https://benchlm.ai), retrieved 2026-09-28 | Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. |
| OSWorld-Verified | 83% | Lab | Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. |
| GDPval-AA v2 | 1421 | Lab | economically valuable knowledge work (v2, re-based Elo) |
| threejseval | 1349 | [threejseval](https://threejseval.com/ranking) | Every model gets the same prompt — "the Eiffel Tower", "a glass fishbowl", "a robot arm picking toys into a box" — and builds a 3D scene in Three.js. Real people then see two scenes side by side, names hidden, and vote for the one they prefer. The votes become a chess-style Elo rating on threejseval.com, averaged across all the prompts. It measures whether the scene looks and moves right to a human eye, not whether the code passes a test. Higher is better. |

## About Gemini 3.6 Flash

Gemini 3.6 Flash was announced on July 21, 2026 alongside Gemini 3.5 Flash-Lite and Gemini 3.5 Flash Cyber. Google emphasized more efficient reasoning for coding agents and professional tasks, including software engineering and machine-learning work.

The release built on the Flash line's focus on response speed and serving cost while improving Google's reported agent evaluations. It supplied the general-purpose model in that announcement; Flash-Lite targeted lighter workloads, while Flash Cyber was introduced for defensive security work.

## Questions and answers

### When was Gemini 3.6 Flash released?

Gemini 3.6 Flash was released by Google on Tuesday, Jul 21 2026.

### Who made Gemini 3.6 Flash?

Gemini 3.6 Flash was built by Google. Builds the Gemini family of models through Google DeepMind. Integrates AI across Google products.

### How much does Gemini 3.6 Flash cost?

Gemini 3.6 Flash costs $0.75 per million input tokens and $3.75 per million output tokens through the Google API. Cached input is $0.075 per million tokens. Output prices include reasoning tokens. Rates are pay-as-you-go API prices verified against Google's published pricing on October 19, 2026.

### What benchmark scores did Gemini 3.6 Flash get?

Gemini 3.6 Flash reports 12 tracked benchmark scores — BullshitBench v2: 39%; Gray Swan IPI (k = 1): 7.3%; Gray Swan IPI (k = 10): 32.2%; Gray Swan IPI (k = 15): 37.3%; ProgramBench: 0.5%; DeepSWE 1.1: 49%; MLE-Bench: 63.9%; BU Bench: 68%; ARC-AGI-2: 60.4%; OSWorld-Verified: 83%; GDPval-AA v2: 1421; threejseval: 1349. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data. It holds the best score among all models tracked here on MLE-Bench.

### What is the context window of Gemini 3.6 Flash?

Gemini 3.6 Flash has a context window of 1M. Providers may specify separate input and output token limits.

### Is Gemini 3.6 Flash open source?

No. Gemini 3.6 Flash is a closed model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms.

### What came before and after Gemini 3.6 Flash?

Google's previous tracked release was Gemini 3.5 Flash on May 19 2026, 63 days earlier. It was followed by Gemini 3.5 Flash-Lite on Jul 21 2026.


---

Canonical page: https://aireleasetracker.com/model/google/gemini-3.6-flash
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.