Tibor Blaho
@btibor91.blaho.me
Lead Engineer at AIPRM.com and LinkResearchTools.com, building AIPRM for ChatGPT & Claude
Pinned
Tibor Blaho
@btibor91.blaho.me
· Jan 9
Did you know there's an extension for ChatGPT & Claude called AIPRM that helps you use and customize prompts and share them with a team?
If you'd like to support my work, this offer is still available for a few more hours (prompt library and toolbox for ChatGPT & Claude) - thank you!
December 1, 2025 at 8:22 AM
If you'd like to support my work, this offer is still available for a few more hours (prompt library and toolbox for ChatGPT & Claude) - thank you!
Did you notice that earlier Claude models had the date in the name match launch day pretty closely, but now Haiku 4.5 (October 1 vs October 15) and Opus 4.5 (November 1 vs November 24) introduce a 2 to 3 week gap between version date and release
November 30, 2025 at 4:46 PM
Did you notice that earlier Claude models had the date in the name match launch day pretty closely, but now Haiku 4.5 (October 1 vs October 15) and Opus 4.5 (November 1 vs November 24) introduce a 2 to 3 week gap between version date and release
What happened with OpenAI and Anthropic this week (Week 48, 2025)
November 30, 2025 at 7:16 AM
What happened with OpenAI and Anthropic this week (Week 48, 2025)
ChatGPT Android app 1.2025.329 beta includes new references to an "ads feature" with "bazaar content", "search ad" and "search ads carousel"
November 29, 2025 at 10:24 AM
ChatGPT Android app 1.2025.329 beta includes new references to an "ads feature" with "bazaar content", "search ad" and "search ads carousel"
If you use ChatGPT or Claude daily, this is your sign to try AIPRM (prompts and more for ChatGPT & Claude) this Black Friday
November 28, 2025 at 8:02 AM
If you use ChatGPT or Claude daily, this is your sign to try AIPRM (prompts and more for ChatGPT & Claude) this Black Friday
ChatGPT web app has a new "ImageGenV2Banner"
"Make images with ChatGPT - Feature your friends, faster image generation, and more consistent editing."
"Make images with ChatGPT - Feature your friends, faster image generation, and more consistent editing."
November 26, 2025 at 10:40 AM
ChatGPT web app has a new "ImageGenV2Banner"
"Make images with ChatGPT - Feature your friends, faster image generation, and more consistent editing."
"Make images with ChatGPT - Feature your friends, faster image generation, and more consistent editing."
Preview of the new directory of apps in the ChatGPT web app (probably a work in progress)
November 25, 2025 at 10:30 PM
Preview of the new directory of apps in the ChatGPT web app (probably a work in progress)
ChatGPT Voice now works right inside the chat interface you already use every day with answers spoken alongside streamed text, image search results, and widgets in the same chat thread
November 25, 2025 at 8:26 PM
ChatGPT Voice now works right inside the chat interface you already use every day with answers spoken alongside streamed text, image search results, and widgets in the same chat thread
Black Forest Labs released FLUX.2 - introducing multi-reference support for up to 10 images with best-in-class character and style consistency, improved photorealism with greater detail and sharper textures, and reliable complex typography rendering
November 25, 2025 at 5:39 PM
Black Forest Labs released FLUX.2 - introducing multi-reference support for up to 10 images with best-in-class character and style consistency, improved photorealism with greater detail and sharper textures, and reliable complex typography rendering
The recording of the next TED Talk from TEDAI Vienna 2025 has been published
November 25, 2025 at 11:04 AM
The recording of the next TED Talk from TEDAI Vienna 2025 has been published
OpenAI launched shopping research in ChatGPT, a new feature that creates personalized buyer's guides, added 6 styles to Sora, and released new MCP connectors
November 24, 2025 at 10:37 PM
OpenAI launched shopping research in ChatGPT, a new feature that creates personalized buyer's guides, added 6 styles to Sora, and released new MCP connectors
Anthropic released Claude Opus 4.5 (claude-opus-4-5-20251101) as their smartest model at $5/$25 per million tokens with top performance for coding, agents, and computer use, launched new beta features for developers, and expanded Claude for Chrome and Claude for Excel to more users
November 24, 2025 at 9:08 PM
Anthropic released Claude Opus 4.5 (claude-opus-4-5-20251101) as their smartest model at $5/$25 per million tokens with top performance for coding, agents, and computer use, launched new beta features for developers, and expanded Claude for Chrome and Claude for Excel to more users
ChatGPT Android app version 1.2025.322 beta has mentions of a new "Personal shopper" (shopping agent)
"I can help you discover products, get ideas, or use personal shopper to research big purchases and find great deals."
"I can help you discover products, get ideas, or use personal shopper to research big purchases and find great deals."
November 24, 2025 at 8:59 AM
ChatGPT Android app version 1.2025.322 beta has mentions of a new "Personal shopper" (shopping agent)
"I can help you discover products, get ideas, or use personal shopper to research big purchases and find great deals."
"I can help you discover products, get ideas, or use personal shopper to research big purchases and find great deals."
This week's OpenAI and Anthropic updates (plus Google's Gemini 3 - Week 47, 2025)
OpenAI released GPT-5.1-Codex-Max and GPT-5.1 Pro, launched free ChatGPT for Teachers, added custom MCP connectors to company knowledge
OpenAI released GPT-5.1-Codex-Max and GPT-5.1 Pro, launched free ChatGPT for Teachers, added custom MCP connectors to company knowledge
November 23, 2025 at 3:02 PM
This week's OpenAI and Anthropic updates (plus Google's Gemini 3 - Week 47, 2025)
OpenAI released GPT-5.1-Codex-Max and GPT-5.1 Pro, launched free ChatGPT for Teachers, added custom MCP connectors to company knowledge
OpenAI released GPT-5.1-Codex-Max and GPT-5.1 Pro, launched free ChatGPT for Teachers, added custom MCP connectors to company knowledge
New Claude web app changes
- new notice for "compaction status" ("Compacting our conversation so we can keep chatting. This takes about 1-2 minutes.")
- new "crimson lighthouse marble nux" banner announcement for the model selector dropdown with a new experiment "blue velvet sofa"
- new notice for "compaction status" ("Compacting our conversation so we can keep chatting. This takes about 1-2 minutes.")
- new "crimson lighthouse marble nux" banner announcement for the model selector dropdown with a new experiment "blue velvet sofa"
November 22, 2025 at 12:57 PM
New Claude web app changes
- new notice for "compaction status" ("Compacting our conversation so we can keep chatting. This takes about 1-2 minutes.")
- new "crimson lighthouse marble nux" banner announcement for the model selector dropdown with a new experiment "blue velvet sofa"
- new notice for "compaction status" ("Compacting our conversation so we can keep chatting. This takes about 1-2 minutes.")
- new "crimson lighthouse marble nux" banner announcement for the model selector dropdown with a new experiment "blue velvet sofa"
ChatGPT now shows more images from the web in answers for topics like people, places, and products, available globally for all plans on web, iOS, and Android when using GPT 5.1
November 22, 2025 at 10:22 AM
ChatGPT now shows more images from the web in answers for topics like people, places, and products, available globally for all plans on web, iOS, and Android when using GPT 5.1
Less than 2 days from US to Austria with UPS this time (compared to 14 days last time because of delays due to bureaucracy), not sure if Logan brought it himself - but thank you, friends at Google!
PS: Gemini 3 Pro is a really large model given the 3XL size I guess
PS: Gemini 3 Pro is a really large model given the 3XL size I guess
November 21, 2025 at 10:57 AM
Less than 2 days from US to Austria with UPS this time (compared to 14 days last time because of delays due to bureaucracy), not sure if Logan brought it himself - but thank you, friends at Google!
PS: Gemini 3 Pro is a really large model given the 3XL size I guess
PS: Gemini 3 Pro is a really large model given the 3XL size I guess
Google introduced Nano Banana Pro (Gemini 3 Pro Image) as their new state-of-the-art image generation and editing model built on Gemini 3 Pro with a 1M token context window and 64K token output that uses Gemini's reasoning and real-world knowledge
November 20, 2025 at 3:24 PM
Google introduced Nano Banana Pro (Gemini 3 Pro Image) as their new state-of-the-art image generation and editing model built on Gemini 3 Pro with a 1M token context window and 64K token output that uses Gemini's reasoning and real-world knowledge
The Claude web app has a new spotlight announcement with the codename "octopus_owl_obsidian" (currently hidden and redacted)
November 20, 2025 at 9:58 AM
The Claude web app has a new spotlight announcement with the codename "octopus_owl_obsidian" (currently hidden and redacted)
OpenAI released GPT-5.1-Codex-Max as the new default coding model that can work for hours across multiple context windows through automatic compaction, seven days after promising more capable agentic and coding models in the weeks ahead in their GPT-5.1 announcement
November 19, 2025 at 11:33 PM
OpenAI released GPT-5.1-Codex-Max as the new default coding model that can work for hours across multiple context windows through automatic compaction, seven days after promising more capable agentic and coding models in the weeks ahead in their GPT-5.1 announcement
Google introduced Gemini 3 as their most intelligent model and a new family of AI models combining state-of-the-art reasoning with multimodal understanding and powerful agentic capabilities
November 18, 2025 at 8:27 PM
Google introduced Gemini 3 as their most intelligent model and a new family of AI models combining state-of-the-art reasoning with multimodal understanding and powerful agentic capabilities
OpenAI Technical Staff Member Fabian Ponce presented at KubeCon+CloudNativeCon North America 2025 in Atlanta how his applied observability team returned about 30,000 CPU cores to their Kubernetes clusters processing nearly 10 petabytes of logs per day by disabling a single function in Fluent Bit
November 16, 2025 at 5:09 PM
OpenAI Technical Staff Member Fabian Ponce presented at KubeCon+CloudNativeCon North America 2025 in Atlanta how his applied observability team returned about 30,000 CPU cores to their Kubernetes clusters processing nearly 10 petabytes of logs per day by disabling a single function in Fluent Bit
Weekly recap of what happened with OpenAI and Anthropic (Week 46, 2025)
November 16, 2025 at 4:36 PM
Weekly recap of what happened with OpenAI and Anthropic (Week 46, 2025)
The recordings of the first TED Talks from TEDAI Vienna 2025 have already been published
November 14, 2025 at 10:59 AM
The recordings of the first TED Talks from TEDAI Vienna 2025 have already been published
OpenAI is piloting group chats (codenamed "calpico") in ChatGPT in Japan, New Zealand, South Korea, and Taiwan for logged-in Free, Go, Plus, and Pro users on mobile and web, where you can invite up to 20 people by sharing a link and anyone in the group can share that link to bring others in
November 14, 2025 at 7:28 AM
OpenAI is piloting group chats (codenamed "calpico") in ChatGPT in Japan, New Zealand, South Korea, and Taiwan for logged-in Free, Go, Plus, and Pro users on mobile and web, where you can invite up to 20 people by sharing a link and anyone in the group can share that link to bring others in