You can try GPT-4.1 (
http://
poe.com/GPT-4.1), mini (
https://
poe.com/GPT-4.1-mini), and nano (
https://
poe.com/GPT-4.1-nano) on web and across all platforms. (3/3)
GENERATIVE AI
-
GPT-4.1 Now Available in Three Versions Across Platforms
By
–
-

OpenAI Launches GPT-4.1 with Enhanced Coding and Long Context
By
–
Now on Poe: GPT-4.1, mini, and nano! GPT-4.1 is OpenAI’s new flagship model with significantly improved coding skills, long context (1M tokens), and improved instruction following. (1/3)
-
GPT-4.1 Mini and Nano Models Launch with Enhanced Performance
By
–
GPT-4.1 mini is smaller, faster, and matches or beats GPT-4o in intelligence and vision-related tasks. Rounding out the new bots, GPT-4.1 nano stands as their fastest and cheapest model yet, with the ability to handle 1M tokens as well. (2/3)
-

Gemini 2.5 Pro Ranks #1 on LMArena with Google Search
By
–
Gemini 2.5 Pro ranks #1 on the LMArena search arena (when powered by grounding with Google Search) right out of the box!
-
AI Murder Report: Critical Slides 38-41 Analysis
By
–
Hello Internet I’d like to report a murder. (See particularly slides 38-41.)
-
Try Windsurf AI: Free GPT-4.1 for a Week of Coding
By
–
Thinking about vibe coding something fun this weekend?
— Romain Huet (@romainhuet) 15 avril 2025
It’s a great time to give @windsurf_ai a spin—free GPT‑4.1 for 7 days! 🏄 https://t.co/6u9OmzCX2zThinking about vibe coding something fun this weekend? It’s a great time to give @windsurf_ai a spin—free GPT‑4.1 for 7 days!
-

OpenAI Launches GPT-4.1 Using Scale AI Benchmark
By
–
Congrats to @OpenAI on the successful launch of their GPT-4.1 model 🙂 Great to see their team utilize the @scale_AI MultiChallenge benchmark to measure multi-turn instruction following
-
GPT-4.1 Mini and Nano Models Released for Speed and Cost
By
–
GPT‑4.1 mini and nano are also available—optimized for speed and cost. Whether you’re shipping new AI features or vibe coding in your IDE, we’d love to hear how GPT‑4.1 is working for you! More details on the blog:
-
Whisper’s Hallucinations: Understanding ASR Limitations
By
–
We dive into the well defined task of automatic speech recognition (ASR), and describe why OpenAI’s Whisper, which has been integrated into ChatGPT, makes stuff up, or “hallucinates” as it’s called in the industry (bad nomenclature).
-
One-Model-Everything Approach Reduces ASR System Reliability
By
–
Their (and Muskrat and others’) quest to build one-model-for-everything has resulted in less reliable systems than what we’ve had before, even in the well defined task of ASR. As in, historically “hallucinations” weren’t problems in ASR systems!