Changelog

🐛 Bug Fixes

• Bug fixed and improvement for credit usage.

Developed with love and passion ♥️♥️

💬 Chat with Claude 3.7 Sonnet, the latest Anthropic Claude model

Available in Pro/ Pro+ plan. Count as 1 advanced token per message.

If you have any feedback or suggestions, please let us know. Contact us at info@bytelify.com or Threads / Instagram

Developed with love and passion ♥️♥️

💬 Chat with o3 mini started from $6/month

The smartest AI model, OpenAI o3 mini is now available in Pro/ Pro+ plan. Count as 2 tokens per message, same as o1 mini.

💬 Unified AI Search

You can now search with all models, including reasoning models (except DeepSeek R1 due to slow response time). We are looking at the AI Search usage (especially for reasoning models), token cost may subject to change.

💬 Introduce New Default Chat Model: Grok 2 (FREE)

We added Grok 2 to our default chat model, and you can now chat with Grok 2 for FREE.

Here are our most recent chat model list and pricing:

Model	Plan	Token(s)/ message	Description
o3 mini	Pro	2
o1 mini	Pro	2
Claude 3.5 Sonnet	Pro	1
GPT 4o	Pro	1
Gemini 2.0 Flash	Pro	1	CurrentlyFree
Gemini 2.0 Pro	Pro	1
Gemini 2.0 Flash Thinking	Pro	1
Sonar Reasoning	Pro	1	Perplexity's R1
DeepSeek R1 (Official)	Pro	1	CurrentlyFree
DeepSeek R1	Pro	1	CurrentlyFree
GPT 4o mini	Free	1
Gemini 2.0 Flash Lite	Free	1
Grok 2	Free	1	Default Model

If you have any feedback or suggestions, please let us know. Contact us at info@bytelify.com or Threads / Instagram

Developed with love and passion ♥️♥️

🖼️ Generate Image with Flux, for FREE!

• Image Generation is now available for all users, for FREE! New Models Added:
• Flux Schnell (free tier)
• Flux Dev (pro tier)

🖼️ Image Generation Price Model Optimized

For free users, you can generate 20/mo on Flux Schnell, and other basic image models in the future. For Pro+ users can generate 250/mo on Pro image models, while Pro users can generate 100/mo. Pro image models include:

• Flux Dev
• Dall-E 3 (4x token)

Other advantages:

• Pro users also generate higher quality and more detailed images in basic models.
• Pro users can generate 2 or 4 images at once.

Developed with love and passion ♥️♥️

📝 New Feature, AI Search 🌐

• AI Search is a new feature base on Chat, that allows you to search the web for latest information.
• Try it out by clicking 🌐 icon in the chat input.
• AI Search is free & available for all users.
• AI Search is available in a limited models, we recommend using Gemini 2.0 Flash for better performance.

💬 Chat with Gemini 2.0 Series

As all Gemini 2.0 models are launched in 5/2/2025, they are now also available in Bytelify, including:

• Gemini 2.0 Flash Lite
• Gemini 2.0 Flash
• Gemini 2.0 Pro
• Gemini 2.0 Flash Thinking
• Gemini 1.5 Flash is no longer available, replaced by Gemini 2.0 Flash Lite.
• Gemini 1.5 Pro is no longer available, replaced by Gemini 2.0 Pro.
• Gemini 2.0 Flash Lite free for all users, count as basic tokens.
• Gemini 2.0 Flash currently free for all users in a limited capacity & time period, count as basic tokens.
• Gemini 2.0 Pro and Gemini 2.0 Flash Thinking are pro models, count as advanced tokens.

All DeepSeek R1 related models (DeepSeek R1, DeepSeek R1 (Llama), Sonar Reasoning) may changed to pro model without early notice.

Developed with love and passion ♥️♥️

💬 Chat with Sonar Reasoning, a reasoning model from the AI serach leader, Perplexity

• Sonar Reasoning is now available in Bytelify, based on Perplexity's DeepSeek R1 model.
• 100% no data will be sent to DeepSeek server as the provider is Perplexity.
• Currently free for all users, count as basic tokens.

All DeepSeek R1 related models (DeepSeek R1, DeepSeek R1 (Llama), Sonar Reasoning) may changed to pro model without early notice.

Developed with love and passion ♥️♥️

💬 ✏️ Say Hello to System Prompt!

• Add support for System Prompt in Chat Page.
• Supported models will pop up a ✏️ to add system prompt.
• When you are applying system prompt, a blue box will show.

💬 Chat with DeepSeek R1 (Llama), experience 10x faster response time than Official Model

• DeepSeek R1 (Llama) is now available in Bytelify, based on DeepSeek R1 Distilled model.
• 100% no data will be sent to DeepSeek server as the provider is Groq.
• Currently free for all users, count as basic tokens.
• DeepSeek R1 Official model may not be fully optimized, expect some slow response.

💬 Pro Model Added: Gemini 2.0 Flash Reasoning, Gemini 2.0 Flash

• Unlock new, fast and reasoning model from Google.
• Currently available for pro and pro+ users, count as advanced tokens.

Developed with love and passion ♥️♥️

😆 UI changes, 🗞️ add News page

UI changes:

• Redesign Footer for supporting News page
• Some UI changes

News:

• Added News, which displays the latest AI news and updates.
• News are generated dynamically by AI.

Developed with love and passion ♥️♥️

💬 Chat with DeepSeek R1, a Chinese Reasoning AI Model

• DeepSeek R1, a Chinese reasoning AI model, is now available in Bytelify.
• Overall performance: Better than OpenAI o1 Mini, but worse than OpenAI o1.
• Currently free for all users, count as basic tokens.

Developed with love and passion ♥️♥️

🐛 Bug Fixes

• Bug fixed and improvement for user profile, also added some caffeine.

Developed with love and passion ♥️♥️

💬 Chat with different providers

• OpenAI: o1 Mini, GPT 4o, GPT 4o Mini
• Anthropic: Claude 3.5 Sonnet
• [BETA] Google Gemini: 2.0 Flash, 1.5 Pro, 1.5 Flash
• [BETA] Meta: Llama 3.2 Vision, 3.3

📸 Image Generation currently supported provider

• OpenAI: Dall E 3

🚀 Subscription plans roll out

• Free: 300 basic tokens per month
• Pro: Unlimited basic tokens, 400 advanced tokens per month
• Pro+: Unlimited basic tokens, 1000 advanced tokens per month

Developed with love and passion ♥️♥️