# Google API Models, Pricing and Capabilities

Canonical HTML page: [Google API Models, Pricing and Capabilities](<https://impossibl.com/google>)
Markdown alternative: [Google API Models, Pricing and Capabilities as Markdown](<https://impossibl.com/google.md>)

Google's Gemini models are natively multimodal, taking text, images, audio, and video, with Pro models for reasoning and Flash models tuned for latency and cost.

Google API and models: 23 models on Impossibl, including Gemini 4 Hydrogen, Gemini Decision 1, and Gemini 4 Carbon. Compare pricing per million tokens, context windows, and capabilities between different Google models.

Select two or three models to [compare specifications and API costs](<https://impossibl.com/compare.md>). Model types and billing units are not interchangeable.

## At a glance

- models: 23
- modalities: takes text, image, audio and video in, gives text, image, embedding, audio and typed answers out
- latest release: Gemini 4 Hydrogen (Nov 12, 2026)

## Models

### [Gemini 2.5 Flash](<https://impossibl.com/google/gemini-2.5-flash>)

Gemini 2.5 Flash is Google's Flash-tier workhorse in the Gemini 2.5 family, positioned as its price-performance choice for low-latency, high-volume tasks that still need reasoning. It carries the family's built-in thinking capability, so it can reason before responding across coding, mathematics, and scientific work at Flash speed. Its published context window is 1,000,000 tokens. The advertised input rate is $0.30 per million input tokens. The advertised output rate is $2.50 per million output tokens.

- API identifier: `google/gemini-2.5-flash`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.30 per million input tokens; $2.50 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-2.5-flash.md>)

### [Gemini 2.5 Flash-Lite](<https://impossibl.com/google/gemini-2.5-flash-lite>)

Gemini 2.5 Flash-Lite is the lightweight, fastest, and most budget-friendly multimodal model in Google's Gemini 2.5 family, built for ultra-low-latency, cost-sensitive workloads. It trades some capability for throughput, with faster token generation than Gemini 2.5 Flash while keeping the family's optional reasoning mode. Its published context window is 1,000,000 tokens. The advertised input rate is $0.10 per million input tokens. The advertised output rate is $0.40 per million output tokens.

- API identifier: `google/gemini-2.5-flash-lite`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.10 per million input tokens; $0.40 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-2.5-flash-lite.md>)

### [Gemini 2.5 Pro](<https://impossibl.com/google/gemini-2.5-pro>)

Gemini 2.5 Pro is Google's Pro-tier model in the Gemini 2.5 family, built for advanced reasoning, coding, mathematics, and scientific tasks. It was Google's first generally available Pro model with built-in thinking, reasoning through a response before answering, and remains the deepest reasoning and coding option in the 2.5 generation. Its published context window is 1,000,000 tokens. The advertised input rate is $1.25 per million input tokens. The advertised output rate is $10.00 per million output tokens. Additional pricing brackets apply when the input exceeds a published context threshold.

- API identifier: `google/gemini-2.5-pro`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $1.25 per million input tokens; $10.00 per million output tokens. Above 200,000 input tokens per request, context-tier rates apply to the whole request.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-2.5-pro.md>)

### [Gemini 3 Pro Image](<https://impossibl.com/google/gemini-3-pro-image>)

Gemini 3 Pro Image, also called Nano Banana Pro, is Google's image generation model for complex visual compositions. It uses reasoning to interpret detailed text prompts and produce images with precise layouts and readable lettering. Its published context window is 131,072 tokens. The advertised input rate is $2.00 per million input tokens. The advertised output rate is $120.00 per million image output tokens. Text and thinking output costs $12.00 per million tokens.

- API identifier: `google/gemini-3-pro-image`
- Creator: Google; model type: Image generation.
- Context window: 131,072 tokens; tool calling: Not supported.
- Inputs: text; outputs: image.
- Base API pricing (USD): $2.00 per million input tokens; $120.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3-pro-image.md>)

### [Gemini 3.1 Flash Image](<https://impossibl.com/google/gemini-3.1-flash-image>)

Gemini 3.1 Flash Image, also called Nano Banana 2, is Google's general-purpose image generation model. It combines prompt understanding with text rendering for creating illustrations, diagrams, and other visual assets. Its published context window is 65,536 tokens. The advertised input rate is $0.50 per million input tokens. The advertised output rate is $60.00 per million image output tokens. Text and thinking output costs $3.00 per million tokens.

- API identifier: `google/gemini-3.1-flash-image`
- Creator: Google; model type: Image generation.
- Context window: 65,536 tokens; tool calling: Not supported.
- Inputs: text; outputs: image.
- Base API pricing (USD): $0.50 per million input tokens; $60.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.1-flash-image.md>)

### [Gemini 3.1 Flash-Lite](<https://impossibl.com/google/gemini-3.1-flash-lite>)

Gemini 3.1 Flash-Lite is Google's generally available Flash-Lite model in the Gemini 3 family, optimized for speed and cost on high-volume workloads. It accepts text, image, video, audio, and PDF input and is designed for lightweight agentic tasks where a larger model would be overkill. Its published context window is 1,000,000 tokens. The advertised input rate is $0.25 per million input tokens. The advertised output rate is $1.50 per million output tokens.

- API identifier: `google/gemini-3.1-flash-lite`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.25 per million input tokens; $1.50 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.1-flash-lite.md>)

### [Gemini 3.1 Flash Lite Image](<https://impossibl.com/google/gemini-3.1-flash-lite-image>)

Gemini 3.1 Flash Lite Image, also called Nano Banana 2 Lite, is Google's speed-focused image generation model. It turns text prompts into images for rapid visual exploration and applications that generate many assets. Its published context window is 65,536 tokens. The advertised input rate is $0.25 per million input tokens. The advertised output rate is $30.00 per million image output tokens. Text and thinking output costs $1.50 per million tokens.

- API identifier: `google/gemini-3.1-flash-lite-image`
- Creator: Google; model type: Image generation.
- Context window: 65,536 tokens; tool calling: Not supported.
- Inputs: text; outputs: image.
- Base API pricing (USD): $0.25 per million input tokens; $30.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.1-flash-lite-image.md>)

### [Gemini 3.1 Flash TTS (preview)](<https://impossibl.com/google/gemini-3.1-flash-tts-preview>)

Gemini 3.1 Flash TTS (preview) is Google's text-to-speech model in the Gemini 3.1 family, generating natural, low-latency speech from text prompts. It adds expressive audio tags for precise narration control alongside steerable prompts and improved naturalness, controllability, and multilingual support over earlier Gemini TTS models. The advertised input rate is $1.00 per million input tokens. The advertised output rate is $20.00 per million output tokens.

- API identifier: `google/gemini-3.1-flash-tts-preview`
- Creator: Google; model type: Text to speech.
- Context window: Not published; tool calling: Not published.
- Inputs: text; outputs: audio.
- Base API pricing (USD): $1.00 per million input tokens; $20.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.1-flash-tts-preview.md>)

### [Gemini 3.1 Pro (preview)](<https://impossibl.com/google/gemini-3.1-pro-preview>)

Gemini 3.1 Pro (preview) is Google's Pro-tier frontier reasoning model, offered in preview for complex problem-solving, agentic coding, and data synthesis. Over Gemini 3 Pro it delivers a step forward in core reasoning, more reliable agentic behaviour, and leaner token usage. Its published context window is 1,000,000 tokens. The advertised input rate is $2.00 per million input tokens. The advertised output rate is $12.00 per million output tokens. Additional pricing brackets apply when the input exceeds a published context threshold.

- API identifier: `google/gemini-3.1-pro-preview`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $2.00 per million input tokens; $12.00 per million output tokens. Above 200,000 input tokens per request, context-tier rates apply to the whole request.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.1-pro-preview.md>)

### [Gemini 3.5 Flash](<https://impossibl.com/google/gemini-3.5-flash>)

Gemini 3.5 Flash is a Flash-tier multimodal model from Google, launched as its workhorse for agentic and coding tasks and now positioned for high-throughput work. It brought near-Pro coding and reasoning to the Flash tier and is optimized for coding proficiency and parallel agentic execution. Its published context window is 1,000,000 tokens. The advertised input rate is $1.50 per million input tokens. The advertised output rate is $9.00 per million output tokens.

- API identifier: `google/gemini-3.5-flash`
- Creator: Google; model type: Text generation.
- Context window: 1,000,000 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $1.50 per million input tokens; $9.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.5-flash.md>)

### [Gemini 3.5 Flash-Lite](<https://impossibl.com/google/gemini-3.5-flash-lite>)

Gemini 3.5 Flash-Lite is Google's fastest and most cost-effective model in the Gemini 3.5 family, built for high-throughput execution and high-volume automation. It is pitched as a low-latency subagent option with upgraded agentic capabilities, suited to focused tasks delegated within larger multi-agent workflows. Its published context window is 1,048,576 tokens. The advertised input rate is $0.30 per million input tokens. The advertised output rate is $2.50 per million output tokens.

- API identifier: `google/gemini-3.5-flash-lite`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.30 per million input tokens; $2.50 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.5-flash-lite.md>)

### [Gemini 3.6 Flash](<https://impossibl.com/google/gemini-3.6-flash>)

Gemini 3.6 Flash is a Flash-tier model from Google for coding, agentic workflows, and web and app development, balancing speed with multimodal capability. It improved token efficiency and agentic planning over Gemini 3.5 Flash at a lower price point, producing polished outputs with fewer unnecessary edits. Its published context window is 1,048,576 tokens. The advertised input rate is $0.75 per million input tokens. The advertised output rate is $3.75 per million output tokens.

- API identifier: `google/gemini-3.6-flash`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.75 per million input tokens; $3.75 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.6-flash.md>)

### [Gemini 3.7 Flash](<https://impossibl.com/google/gemini-3.7-flash>)

Gemini 3.7 Flash is a Flash-tier multimodal model from Google for coding, agentic workflows, and reliable multi-step execution, now superseded by Gemini 3.8 Flash. Google positioned it as its workhorse for coding and agents at launch, with substantial gains over Gemini 3.6 Flash in software engineering and web development. Its published context window is 1,048,576 tokens. The advertised input rate is $0.75 per million input tokens. The advertised output rate is $3.75 per million output tokens.

- API identifier: `google/gemini-3.7-flash`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.75 per million input tokens; $3.75 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.7-flash.md>)

### [Gemini 3.8 Flash](<https://impossibl.com/google/gemini-3.8-flash>)

Gemini 3.8 Flash is Google's current Flash-tier workhorse model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows. It keeps the speed and cost of Gemini 3.7 Flash while working harder on difficult tasks through extra reasoning and iterative tool calls, plus a cyber-defense variant for vetted defenders. Its published context window is 1,048,576 tokens. The advertised input rate is $0.75 per million input tokens. The advertised output rate is $3.75 per million output tokens.

- API identifier: `google/gemini-3.8-flash`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $0.75 per million input tokens; $3.75 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.8-flash.md>)

### [Gemini 3.8 Flash TTS](<https://impossibl.com/google/gemini-3.8-flash-tts>)

Gemini 3.8 Flash TTS is Google's text-to-speech model in the Gemini 3.8 family, turning text prompts into expressive 24 kHz speech across 30 prebuilt voices. It is steerable in natural language, so style, pace, accent and tone follow the prompt rather than fixed presets. The advertised input rate is $0.50 per million input tokens. The advertised output rate is $9.00 per million output tokens.

- API identifier: `google/gemini-3.8-flash-tts`
- Creator: Google; model type: Text to speech.
- Context window: Not published; tool calling: Not published.
- Inputs: text; outputs: audio.
- Base API pricing (USD): $0.50 per million input tokens; $9.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-3.8-flash-tts.md>)

### [Gemini 4 Argon](<https://impossibl.com/google/gemini-4-argon>)

Gemini 4 Argon is Google's first Gemini 4 model, a frontier system built to sustain careful reasoning across long, multi-step professional work. It is aimed at real-world software engineering, enterprise knowledge work such as legal and finance, and cybersecurity defense, rather than the quick everyday turns the Flash line handles. Its published context window is 1,048,576 tokens. The advertised input rate is $2.00 per million input tokens. The advertised output rate is $10.00 per million output tokens.

- API identifier: `google/gemini-4-argon`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $2.00 per million input tokens; $10.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-4-argon.md>)

### [Gemini 4 Carbon](<https://impossibl.com/google/gemini-4-carbon>)

Gemini 4 Carbon is Google's strongest Gemini 4 model, built on Argon's foundation with extra post-training for deep reasoning and extended verification. It is meant for the hardest multi-step coding, debugging and scientific research work, where it spends more thinking tokens than Argon, which stays the faster choice for everyday engineering. Its published context window is 1,048,576 tokens. The advertised input rate is $2.00 per million input tokens. The advertised output rate is $10.00 per million output tokens.

- API identifier: `google/gemini-4-carbon`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $2.00 per million input tokens; $10.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-4-carbon.md>)

### [Gemini 4 Hydrogen](<https://impossibl.com/google/gemini-4-hydrogen>)

Gemini 4 Hydrogen is Google's fast, cost-efficient Gemini 4 model, the successor to Gemini 3.8 Flash. It is meant for everyday coding and agentic workflows, trading some of the depth of Argon and Carbon for speed and a lower cost per task. Its published context window is 1,048,576 tokens. The advertised input rate is $1.50 per million input tokens. The advertised output rate is $7.50 per million output tokens.

- API identifier: `google/gemini-4-hydrogen`
- Creator: Google; model type: Text generation.
- Context window: 1,048,576 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $1.50 per million input tokens; $7.50 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-4-hydrogen.md>)

### [Gemini Decision 1](<https://impossibl.com/google/gemini-decision-1>)

Gemini Decision 1 is Google's decision model for fast, closed-set judgments: it reads one shared multimodal input (text, images, audio, video or PDFs) and scores up to 1,000 yes/no, multiple-choice and rubric questions in a single forward pass. It returns probabilities and a confidence for each answer instead of generated text, which suits routing, policy checks, tool selection and grading. Its published context window is 262,144 tokens. The advertised input rate is $0.10 per million input tokens. The advertised output rate is $0.00 per million output tokens.

- API identifier: `google/gemini-decision-1`
- Creator: Google; model type: Evaluation.
- Context window: 262,144 tokens; tool calling: Not supported.
- Inputs: text, image, audio, video; outputs: typed answers with probabilities.
- Base API pricing (USD): $0.10 per million input tokens; $0.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-decision-1.md>)

### [Gemini Embedding 2](<https://impossibl.com/google/gemini-embedding-2>)

Gemini Embedding 2 is Google's embedding model for representing content as vectors that capture meaning. Its text embeddings support semantic search, document retrieval, classification, and clustering across languages. Its published context window is 8,192 tokens. The advertised input rate is $0.20 per million input tokens.

- API identifier: `google/gemini-embedding-2`
- Creator: Google; model type: Embeddings.
- Context window: 8,192 tokens; tool calling: Not supported.
- Inputs: text; outputs: embedding.
- Base API pricing (USD): $0.20 per million input tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-embedding-2.md>)

### [Nano Banana 2.1](<https://impossibl.com/google/gemini-nano-banana-2.1>)

Nano Banana 2.1 is Google's Flash-tier image generation and conversational editing model, the successor to Nano Banana 2. It is built for high-volume creative work where visual quality, text rendering and keeping a character consistent across several turns of editing matter. Its published context window is 65,536 tokens. The advertised input rate is $1.50 per million input tokens. The advertised output rate is $30.00 per million image output tokens. Text and thinking output costs $7.50 per million tokens.

- API identifier: `google/gemini-nano-banana-2.1`
- Creator: Google; model type: Image generation.
- Context window: 65,536 tokens; tool calling: Not supported.
- Inputs: text; outputs: image.
- Base API pricing (USD): $1.50 per million input tokens; $30.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-nano-banana-2.1.md>)

### [Gemini Robotics-ER 2 Preview](<https://impossibl.com/google/gemini-robotics-er-2-preview>)

Gemini Robotics-ER 2 Preview is Google's embodied-reasoning vision-language model for robotics, taking text, image, video, and audio input and producing text for task planning. It supports spatial reasoning, video moment finding, agentic code execution, multi-step tool orchestration, and multi-robot coordination, adding caching, code execution, and computer use over Robotics-ER 1.6. Its published context window is 131,072 tokens. The advertised input rate is $1.00 per million input tokens. The advertised output rate is $5.00 per million output tokens.

- API identifier: `google/gemini-robotics-er-2-preview`
- Creator: Google; model type: Text generation.
- Context window: 131,072 tokens; tool calling: Not published.
- Inputs: text, image, audio, video; outputs: text.
- Base API pricing (USD): $1.00 per million input tokens; $5.00 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemini-robotics-er-2-preview.md>)

### [Gemma 4 31B](<https://impossibl.com/google/gemma-4-31b>)

Gemma 4 31B is the largest dense open-weight model in Google's Gemma 4 family, released under Apache 2.0 for advanced reasoning and agentic workflows. Unlike the smaller E2B and E4B variants it takes text and image but not audio, favouring quality over speed. Its published context window is 128,000 tokens. The advertised input rate is $0.99 per million input tokens. The advertised output rate is $1.49 per million output tokens.

- API identifier: `google/gemma-4-31b`
- Creator: Google; model type: Text generation.
- Context window: 128,000 tokens; tool calling: Not published.
- Inputs: text, image; outputs: text.
- Base API pricing (USD): $0.99 per million input tokens; $1.49 per million output tokens.
- [Complete model facts, cache rates, context tiers, and FAQs as Markdown](<https://impossibl.com/google/gemma-4-31b.md>)

## Compare Google models

- [Gemini 2.5 Flash vs Gemini 2.5 Flash-Lite](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-2.5-flash-lite>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-2.5-flash-lite.md>)
- [Gemini 2.5 Flash vs Gemini 2.5 Pro](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-2.5-pro>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-2.5-pro.md>)
- [Gemini 2.5 Flash vs Gemini 3.1 Flash-Lite](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.1-flash-lite>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.1-flash-lite.md>)
- [Gemini 2.5 Flash vs Gemini 3.1 Pro (preview)](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.1-pro-preview>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.1-pro-preview.md>)
- [Gemini 2.5 Flash vs Gemini 3.5 Flash](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.5-flash>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.5-flash.md>)
- [Gemini 2.5 Flash vs Gemini 3.5 Flash-Lite](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.5-flash-lite>); [Markdown comparison](<https://impossibl.com/compare/google/gemini-2.5-flash/google/gemini-3.5-flash-lite.md>)

## Frequently asked questions

### Which Gemini models are available through Impossibl?

The directory lists 23 models from Google available through the Impossibl API, including Gemini 4 Hydrogen, Gemini Decision 1, and Gemini 4 Carbon. Open each model page for its API identifier, capabilities, and current pricing.

### How much do Gemini models cost?

Google model prices vary by model and usage. All rates below are published base API rates in USD. For token-priced models, published input rates start at $0.10 per million input tokens. Published output rates start at $0.00 per million output tokens. Input and output minima may refer to different models. Long-context tiers can change prices for the whole request; check each model's thresholds and rates. Individual model pages include published cache prices and billing units. Applicable fees and workspace-specific pricing can affect the amount paid.

### What is the cheapest Gemini model?

By published input rate, the cheapest Gemini text generation model on Impossibl is Gemini 2.5 Flash-Lite (google/gemini-2.5-flash-lite) at $0.10 per million input tokens and $0.40 per million output tokens. Cheapest by input rate is not always cheapest for a workload; output-heavy or cached requests can favor another model.

### Does Impossibl add a markup to Google pricing?

No. Usage is charged at the provider's list price with 0% usage markup. Prepaid credit purchases include a 5% platform fee, and there is no subscription. The rates on this page are Google's published base rates, and usage comes off your balance when the request finishes.

### What context windows do Gemini models support?

Among Google models with a published limit on Impossibl, the largest context window is 1,048,576 tokens, listed for Gemini 3.5 Flash-Lite, Gemini 3.6 Flash, and Gemini 3.7 Flash. Context limits vary by model and are separate from maximum output limits. Context windows are unpublished for 2 models.

### What inputs and capabilities do Gemini models support?

Available model types include embeddings, evaluation, image generation, text generation, and text to speech. Published input support includes audio, image, text, and video; support varies by model. Tool calling is published as unsupported for 6 models. Tool-calling support is unspecified for 17 models.

### How do I use Gemini models through an API?

Use the exact model identifier from the model page with the Impossibl API. For example, select google/gemini-4-hydrogen with /v1/chat/completions. Endpoint compatibility depends on the selected model. The API quickstart explains how to create an API key and send a request.

## Sources and related links

Prices, capabilities, limits, and availability describe the public Impossibl API offering. An unpublished fact is unknown; a published zero price is zero. Compare matching billing units and check context tiers before estimating costs.

- [Public Impossibl model catalog](<https://api.impossibl.com/v1/models>)
- [Full model directory as JSON](<https://impossibl.com/models.json>)
- [Model directory as Markdown](<https://impossibl.com/models.md>)
- [API quickstart](<https://impossibl.com/docs/agent-quickstart>)
- [Model documentation](<https://impossibl.com/docs/models>)
