Video

Google

Gemini 3.8 Flash

Gemini Flash

Latest Gemini Flash. Reads text, images, audio and video; 1M context.

• Audio & video
• Agentic work
• Long documents

Pricing

$0.75 / $3.75 per 1M tokens, in / out

Context window

1M context

Google

Gemini 3.5 Flash

Gemini Flash

Near-Pro coding and reasoning at Flash cost, with parallel tool use.

• Coding
• Audio & video
• Agentic work

Pricing

$1.5 / $9 per 1M tokens, in / out

Context window

1M context

Google

Gemini 3.6 Flash

Gemini Flash

Flash model for coding and web/app development with fewer stray edits.

• Coding
• Audio & video

Pricing

$0.75 / $3.75 per 1M tokens, in / out

Context window

1M context

Google

Gemini 3.5 Flash Lite

Gemini Flash-Lite

Cheap Gemini for sub-agents that run focused tasks.

• Low cost
• Audio & video
• Fast replies

Pricing

$0.3 / $2.5 per 1M tokens, in / out

Context window

1M context

Google

Gemini 2.5 Pro

Gemini Pro

Gemini with step-by-step thinking for math, science and code.

• Reasoning
• Audio & video
• Long documents

Pricing

$1.25 / $10 per 1M tokens, in / out

Context window

1M context

Google

Gemini 3.1 Flash Lite

Gemini Flash-Lite

Low-latency Gemini for high volume. Reads audio, video and PDF.

• Low cost
• Fast replies
• Audio & video

Pricing

$0.25 / $1.5 per 1M tokens, in / out

Context window

1M context

Google

Gemini 2.5 Flash Lite

Gemini Flash-Lite

Among the cheapest multimodal models here, 1M context.

• Low cost
• Fast replies
• Audio & video

Pricing

$0.1 / $0.4 per 1M tokens, in / out

Context window

1M context

Google

Gemma 4 31B

Gemma 4

Open-weight dense model with image and video input and function calling.

• Open weights
• Low cost
• Images

Pricing

$0.09 / $0.34 per 1M tokens, in / out

Context window

262K context

Google

Gemma 4 26B

Gemma 4

Open-weight MoE close to Gemma 4 31B quality at lower compute.

• Open weights
• Low cost
• Fast replies

Pricing

$0.09 / $0.3 per 1M tokens, in / out

Context window

262K context

Moonshot AI

Kimi K3

Kimi

Large open-weight multimodal model for coding and long agent runs.

• Coding
• Agentic work
• Open weights

Pricing

$1.7 / $8.5 per 1M tokens, in / out

Context window

1M context

Alibaba Qwen

Qwen3.8 Max

Qwen3.8

Largest Qwen, with text, image and video input.

• Reasoning
• Coding
• Audio & video

Pricing

$2 / $6 per 1M tokens, in / out

Context window

1M context

Alibaba Qwen

Qwen3.8 Flash

Qwen3.8

Cheap multimodal Qwen for charts, documents and long videos.

• Low cost
• Audio & video
• Images

Pricing

$0.15 / $0.47 per 1M tokens, in / out

Context window

1M context

Alibaba Qwen

Qwen3.6 Flash

Qwen3.6

Fast Qwen with image and video input and a 1M context.

• Fast replies
• Low cost
• Audio & video

Pricing

$0.19 / $1.13 per 1M tokens, in / out

Context window

1M context

Z.ai

GLM-5V Turbo

GLM-5V

GLM that reads images and video, for vision-based coding.

• Images
• Coding
• Audio & video

Pricing

$1.2 / $4 per 1M tokens, in / out

Context window

203K context

MiniMax

MiniMax M3

MiniMax

Multimodal model with a 1M context for long agent work.

• Audio & video
• Agentic work
• Low cost

Pricing

$0.3 / $1.2 per 1M tokens, in / out

Context window

1M context

Meta

Muse Spark 1.3

Muse Spark

Keeps track of long tasks. Reads text, images, audio, video and PDF.

• Agentic work
• Audio & video
• Long documents

Pricing

$1.25 / $4.25 per 1M tokens, in / out

Context window

1M context

Meta

Muse Spark 1.1

Muse Spark

Multimodal reasoning for agent tasks, 1M context.

• Agentic work
• Audio & video

Pricing

$1.25 / $4.25 per 1M tokens, in / out

Context window

1M context

StepFun

Step 3.7 Flash

Step

Efficient MoE with native image and video understanding.

• Audio & video
• Low cost
• Images

Pricing

$0.2 / $1.15 per 1M tokens, in / out

Context window

262K context