Models

Every model behind one OpenAI-compatible key.

Type

Providers

44 models

Claude Fable 5: the Mythos-class model for complex, long-running planning, coding and vision work.

by anthropic200K context$6.00/M input$7.50$30.00/M output$37.50

OpenAI: GPT-5.6 Pro

−15%

Flagship GPT-5.6 Pro for the hardest agentic and coding tasks.

by openai400K context$4.25/M input$5.00$25.50/M output$30.00

Anthropic: Claude Opus 4.8

−10%

The newest flagship Claude Opus 4.8 — Anthropic's most capable model.

by anthropic200K context$4.50/M input$5.00$22.50/M output$25.00

OpenAI: GPT-5.6

−15%

The refreshed GPT-5.6 workhorse: strong code, reasoning and tool calling, 400K context.

by openai400K context$2.13/M input$2.50$12.75/M output$15.00

Google: Gemini 3.1 Pro

−15%

Gemini 3.1 Pro for complex reasoning and code, 1M context.

by google1M context$1.46/M input$2.00$10.20/M output$12.00

Anthropic: Claude Sonnet 5

−15%

Claude Sonnet 5: the next-gen Sonnet — strong coding and writing at a lower price.

by anthropic200K context$1.69/M input$2.00$8.50/M output$10.00

Google: Gemini 3.5 Flash

−10%

The newest fast Gemini 3.5 — strong reasoning at low latency.

by google1M context$1.35/M input$1.50$8.10/M output$9.00

xAI: Grok 4.5

−15%

xAI's Grok 4.5: a frontier model for coding, agentic tasks and reasoning.

by xai256K context$1.70/M input$2.00$5.10/M output$6.00

OpenAI: GPT-5.6 Mini

−15%

GPT-5.6 Mini: a fast, economical option for high-volume tasks.

by openai400K context$0.85/M input$1.00$5.10/M output$6.00

Anthropic: Claude Haiku 4.5

−20%

The fastest, cheapest Claude for light tasks and high volume.

by anthropic200K context$0.77/M input$1.00$4.00/M output$5.00

Google: Gemini 3.1 Flash Lite

−10%

The cheapest, fastest Gemini for high-volume tasks, 1M context.

by google1M context$0.22/M input$0.25$1.35/M output$1.50

Google: Nano Banana Pro

−10%

The Pro Nano Banana with up to 4K output.

by googlefrom $0.12/image$0.13

Google: Nano Banana 2

−30%

Second-gen Nano Banana: higher quality and detail.

by googlefrom $0.071/image$0.10

ByteDance: Seedream 5 Pro

−16%

ByteDance Seedream 5 Pro: controlled generation, precise editing, realistic portraits and multilingual text.

by bytedancefrom $0.045/image$0.053

ByteDance: Seedream 4.5

−15%

ByteDance Seedream 4.5: improved quality and editing.

by bytedancefrom $0.040/image$0.047

OpenAI: GPT Image 2

−30%

OpenAI image generation and editing with accurate text rendering.

by openaifrom $0.037/image$0.053

Google: Nano Banana Fast

−10%

Google's ultra-fast, cheap image generation.

by googlefrom $0.035/image$0.039

ByteDance: Seedream 5 Lite

−15%

Seedream 5 Lite: a fast, cheap option in the Seedream line.

by bytedancefrom $0.035/image$0.041

Black Forest: FLUX.2 Pro

−15%

FLUX.2 Pro by Black Forest Labs: top photorealism and accurate text.

by black-forestfrom $0.030/image$0.035

xAI: Grok Imagine

−20%

xAI's Grok Imagine: image generation and editing.

by xaifrom $0.030/image$0.037

Alibaba: Wan 2.7

−25%

Alibaba Wan 2.7: image generation and inline editing up to 2K.

by alibabafrom $0.030/image$0.040

Alibaba: Qwen Image

−50%

Alibaba's Qwen-Image: generation and editing with strong in-image text.

by alibabafrom $0.018/image$0.035

Google: Veo 3.1

−20%

Google DeepMind's Veo 3.1: cinematic 1080p video with native audio, from text or a first frame.

by googleJul 23, 2026from $2.55/video$3.20

ByteDance: Seedance 2.0

−18%

ByteDance Seedance 2.0: multimodal video with strong multi-shot consistency, up to 4K.

by bytedanceJul 23, 2026from $0.25/second$0.30

Google: Veo 3.1 Fast

−19%

The fast tier of Veo 3.1 — the same look for a fraction of the price, 1080p with audio.

by googleJul 23, 2026from $0.97/video$1.20

Alibaba: Wan 2.7 Video

−8%

Alibaba Wan 2.7 Video: 720p–1080p clips up to 15 seconds from text or a first frame, with audio.

by alibabaJul 23, 2026from $0.092/second$0.10

Kling: Kling 3.0

Kling 3.0: multi-shot scenes up to 15 seconds, native audio and std/pro/4K modes.

by klingJul 23, 2026from $0.087/second

MiniMax: Hailuo 2.3

−20%

MiniMax Hailuo 2.3: image-to-video with lifelike motion, facial expression and lighting.

by minimaxJul 23, 2026from $0.037/second$0.047

OpenAI: GPT-5.5

−15%

Flagship GPT-5.5 for the hardest agentic and coding tasks.

by openai400K context$4.16/M input$5.00$25.50/M output$30.00

Anthropic: Claude Opus 4.6

−20%

Claude Opus 4.6: top-tier reasoning and autonomous coding.

by anthropic200K context$3.99/M input$5.00$20.00/M output$25.00

Anthropic: Claude Opus 4.7

−20%

Claude Opus 4.7: top-tier reasoning and autonomous coding.

by anthropic200K context$3.99/M input$5.00$20.00/M output$25.00

Anthropic: Claude Opus 4.5

−25%

Claude Opus 4.5 for complex, long-running tasks with few mistakes.

by anthropic200K context$3.74/M input$5.00$18.75/M output$25.00

Anthropic: Claude Sonnet 4.5

−10%

Balanced Claude Sonnet: strong coding and writing at low latency.

by anthropic200K context$2.68/M input$3.00$13.50/M output$15.00

Anthropic: Claude Sonnet 4.6

−10%

Claude Sonnet 4.6: a refreshed balance of speed and quality.

by anthropic200K context$2.68/M input$3.00$13.50/M output$15.00

OpenAI: GPT-5.4

−15%

The GPT-5.4 workhorse: strong code, reasoning and tool calling, 400K context.

by openai400K context$1.49/M input$2.50$12.75/M output$15.00

OpenAI: GPT-5.2

−20%

GPT-5.2: a fast, economical option for high-volume tasks.

by openai400K context$1.40/M input$1.75$11.20/M output$14.00

Google: Gemini 2.5 Pro

−15%

Multimodal Gemini Pro with a 1M context: text, image, audio, video.

by google1M context$1.08/M input$1.25$8.50/M output$10.00

Google: Gemini 3 Pro

−15%

Flagship Gemini 3 Pro for the hardest multimodal tasks.

by google1M context$0.75/M input$0.88$5.25/M output$6.18

xAI: Grok 4.3

−15%

xAI's Grok 4.3: a fast, economical option in the Grok line.

by xai256K context$1.06/M input$1.25$2.13/M output$2.50

Google: Gemini 3 Flash

−10%

Gen-3 fast multimodal Gemini with improved reasoning.

by google1M context$0.45/M input$0.50$2.70/M output$3.00

Google: Gemini 2.5 Flash

−10%

Fast, low-cost Gemini for high-throughput workloads, 1M context.

by google1M context$0.27/M input$0.30$2.25/M output$2.50

Google: Nano Banana

−10%

Google's fast image generation and editing with good control.

by googlefrom $0.035/image$0.039

ByteDance: Seedream V4

−15%

ByteDance Seedream V4: photoreal generation and editing.

by bytedancefrom $0.030/image$0.035

Kling: Kling 2.6

−2%

Kling 2.6: synchronized video and audio — speech, ambience and effects — from text or an image.

by klingJul 23, 2026from $0.069/second$0.070