Pricing
Prices shown are what you pay. No hidden fees, no subscriptions.
AiFuse charges what the provider charges us, plus a margin. That margin is what pays for the workspace — one login for every model, one balance across chat, images and voice. Every price on this page is the price you pay, and every message shows what it cost the moment it arrives.
xAI
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Grok 4.6 | Frontier reasoning model. 500K context. | $3.20 | $8.40 |
| Grok 4.5 | Frontier reasoning model. 200K context. | $3.20 | $8.40 |
| Grok BuildFree trial | Fast coding model for agentic workflows. 256K context. | $1.60 | $2.80 |
| Grok 4.3Free trial | xAI's most powerful model with extended reasoning capability. | $2.00 | $3.50 |
Anthropic
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Claude Fable 5 | Anthropic's most advanced model for the hardest reasoning and long-horizon agentic work. | $16.00 | $70.00 |
| Claude Opus 5 | Anthropic's latest flagship for deep reasoning and long-horizon agentic coding. 1M context. | $8.00 | $35.00 |
| Claude Sonnet 5Free trial | Balanced, versatile workhorse for daily use and scaled production. | $4.80 | $21.00 |
| Claude Haiku 4.5Free trial | Fastest and most affordable for quick answers | $1.60 | $7.00 |
| Claude Opus 4.8 | Highly autonomous flagship for complex reasoning, long-horizon agentic coding, and knowledge work. | $8.00 | $35.00 |
DeepSeek
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| DeepSeek V4 Pro | Flagship 1.6T-parameter MoE reasoning for complex coding, multi-step analysis, and long-horizon agentic work, at frontier quality. | $2.11 | $5.54 |
| DeepSeek V4 FlashFree trial | Fast and affordable with strong coding and analysis. 128K context. | $0.70 | $1.85 |
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Gemini 3.1 ProFree trial | Google's flagship reasoning model with strong multimodal and long-context performance. | $3.20 | $16.80 |
| Gemini 3.6 FlashFree trial | Google's headline Flash model. Fast, high quality, 1M context. | $2.40 | $10.50 |
Moonshot
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Kimi K3Free trial | Moonshot's reasoning model with a 1M-token context and native vision. Strong multi-step problem solving. | $4.80 | $21.00 |
| Kimi K2.7 CodeFree trial | Moonshot's coding-specialist Kimi with a 1M-token context. Tuned for code generation, editing, and repo-scale reasoning. | $1.52 | $5.60 |
OpenAI
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| GPT-5.5 Pro | OpenAI's specialist GPT-5.5 Pro model for the hardest, highest-stakes reasoning. | $48.00 | $252.00 |
| GPT-5.6 Sol | OpenAI's top-tier GPT-5.6 model for the most complex reasoning and agentic tasks. | $6.40 | $28.00 |
| GPT-5.6 Terra | OpenAI's balanced GPT-5.6 model for everyday complex tasks and reasoning. | $3.20 | $16.80 |
| GPT-5.6 LunaFree trial | OpenAI's fast, cost-efficient GPT-5.6 model for high-volume tasks. | $0.32 | $1.68 |
| GPT-5.5 | Flagship reasoning model. 1.05M context, 128K output. Highest-end OpenAI tier. | $8.00 | $42.00 |
| GPT-5.4 | OpenAI's flagship model for complex tasks and advanced reasoning. | $4.00 | $21.00 |
| GPT-5.4 NanoFree trial | Ultra-fast and ultra-cheap for simple and repetitive tasks. | $0.32 | $1.75 |
Perplexity
| Model | Description | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Perplexity Sonar Pro | Advanced web-grounded search with deeper reasoning and more citations. 200K context. | $4.80 | $21.00 |
| Perplexity Sonar Reasoning Pro | Web-grounded reasoning with chain-of-thought and real-time citations. 128K context. | $3.20 | $11.20 |
| Perplexity SonarFree trial | Web-grounded AI search with real-time citations. 127K context. | $1.60 | $1.40 |
Sonar models include a per-request search fee on top of token rates ($0.005–$0.022 depending on search context size). AiFuse covers this fee — your billing reflects the token rates above.
Image Generation
| Model | Description | Price |
|---|---|---|
| GPT Image 2OpenAI | Versatile image generation | $0.07 / image |
| Grok ImaginexAI | Fast AI image generation | $0.03 / image |
| Grok Imagine ProxAI | High-quality AI image generation | $0.07 / image |
| Nano Banana 2Google | Google's fast native multimodal image model. Token-based billing. | $0.09 / image |
| Nano Banana ProGoogle | Google's flagship native multimodal image model. Token-based billing. | $0.18 / image |
| Recraft V4.1Recraft | High-quality raster image generation with strong typography and design control. | $0.05 / image |
| Recraft V4.1 ProRecraft | Premium raster tier for the highest-fidelity image generation. | $0.28 / image |
| Recraft V4.1 Pro VectorRecraft | Premium SVG output for professional, production-ready vector assets. | $0.41 / image |
| Recraft V4.1 UtilityRecraft | Fast, cost-efficient raster generation for high-volume needs. | $0.05 / image |
| Recraft V4.1 Utility ProRecraft | Higher-fidelity utility raster tier for demanding volume work. | $0.28 / image |
| Recraft V4.1 VectorRecraft | True SVG output for logos, icons, and infinitely scalable illustrations. | $0.11 / image |
Video Generation
| Model | Description | Price |
|---|---|---|
| Grok VideoxAI | AI video generation | $0.08 / second |
Web Search
$0.0075 per search — only when a message needs the web.
All prices in USD per 1M tokens. Most messages cost a fraction of a cent to a few cents.
Weighing us against the alternatives? We wrote an honest comparison of four ways to run Claude and GPT together, including the ones that cost less than us.