# Model Discussions **URL:** https://community.pickaxe.co/c/llm-ai-model-discussions/12.md [Latest](https://community.pickaxe.co/latest.md) · [Categories](https://community.pickaxe.co/categories.md) · [Tags](https://community.pickaxe.co/tags.md) --- ## [About the Model Discussions category](https://community.pickaxe.co/t/about-the-model-discussions-category/7803)
**Author:** [@abhi](https://community.pickaxe.co/u/abhi)\ **Replies:** 0\ **Last updated:** [December 18, 2025, 4:20pm UTC](https://community.pickaxe.co/t/about-the-model-discussions-category/7803 "2025-12-18T16:20:58Z")
Discuss LLMs used in Pickaxe. Compare models, costs, strengths, limitations, and real-world builder experiences. Why should people use this category? What is it for? This category is a dedicated space to talk about lar… --- ## [GPT Image 1.5 vs Google Nanobanana Pro](https://community.pickaxe.co/t/gpt-image-1-5-vs-google-nanobanana-pro/7833)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 2\ **Last updated:** [March 20, 2026, 12:07am UTC](https://community.pickaxe.co/t/gpt-image-1-5-vs-google-nanobanana-pro/7833 "2026-03-20T00:07:07Z")
My take: GPT Image 1.5 is MUCH better at making non destructive tweaks and changes. GPT Image 1.5 is better at illustrations than Nanobanana. Nanobanana is better at photo realistic images than Nanobanana… --- ## [Grok 4.1 Fast Reasoning Crazy Smart & Crazy Cheap](https://community.pickaxe.co/t/grok-4-1-fast-reasoning-crazy-smart-crazy-cheap/7805)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 2\ **Last updated:** [March 20, 2026, 12:00am UTC](https://community.pickaxe.co/t/grok-4-1-fast-reasoning-crazy-smart-crazy-cheap/7805 "2026-03-20T00:00:35Z")
I really like Grok 4.1 Fast Reasoning. It is my go to model for most tasks. Grok 4.1 Fast is cheaper and faster since it doesn’t have reasoning. Grok Fast is an incredibly affordable model. The reason the Pickaxe shows… --- ## [Claude Drops Prices for Large Contexts](https://community.pickaxe.co/t/claude-drops-prices-for-large-contexts/8229)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 1\ **Last updated:** [March 15, 2026, 8:36am UTC](https://community.pickaxe.co/t/claude-drops-prices-for-large-contexts/8229 "2026-03-15T08:36:06Z")
Exciting news from Anthropic today: They’ve made their massive 1M token context window generally available for Claude Opus 4.6 and Sonnet 4.6 models. If you’re building Pickaxes that deal with hefty datasets, codebases, … --- ## [ChatGPT Is Getting Ads. Pickaxe Just Became More Valuable](https://community.pickaxe.co/t/chatgpt-is-getting-ads-pickaxe-just-became-more-valuable/8049)
**Author:** [@abhi](https://community.pickaxe.co/u/abhi)\ **Replies:** 1\ **Last updated:** [February 20, 2026, 7:27pm UTC](https://community.pickaxe.co/t/chatgpt-is-getting-ads-pickaxe-just-became-more-valuable/8049 "2026-02-20T19:27:16Z")
Quick heads up :eyes: OpenAI is officially testing ads inside ChatGPT. Official announcement: https://openai.com/index/testing-ads-in-chatgpt/ This creates a strong opportunity for Pickaxe builders. You can position P… --- ## [How do you get Claude to make tool calls?](https://community.pickaxe.co/t/how-do-you-get-claude-to-make-tool-calls/7811)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 0\ **Last updated:** [December 18, 2025, 11:06pm UTC](https://community.pickaxe.co/t/how-do-you-get-claude-to-make-tool-calls/7811 "2025-12-18T23:06:10Z")
One of the things we talked about in today’s office hours was tool calling with claude. Claude is the most hesitant to call on tools. So what are your tricks? I’ll start be super explicit in the prompt that you want the… --- ## [Gemini 3 vs GPT 5.2 which do you like better?](https://community.pickaxe.co/t/gemini-3-vs-gpt-5-2-which-do-you-like-better/7808)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 0\ **Last updated:** [December 18, 2025, 7:18pm UTC](https://community.pickaxe.co/t/gemini-3-vs-gpt-5-2-which-do-you-like-better/7808 "2025-12-18T19:18:29Z")
--- ## [Me Talking About LLM Models for Writing](https://community.pickaxe.co/t/me-talking-about-llm-models-for-writing/7807)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 0\ **Last updated:** [December 18, 2025, 5:19pm UTC](https://community.pickaxe.co/t/me-talking-about-llm-models-for-writing/7807 "2025-12-18T17:19:22Z")
There are maybe three of you who would be interested in this video, but in the interest of populating the new Model Discussion board here it is: --- ## [Gemini 3 Flash is Hallucination Central](https://community.pickaxe.co/t/gemini-3-flash-is-hallucination-central/7806)
**Author:** [@thomasumstattd](https://community.pickaxe.co/u/thomasumstattd)\ **Replies:** 1\ **Last updated:** [December 18, 2025, 5:06pm UTC](https://community.pickaxe.co/t/gemini-3-flash-is-hallucination-central/7806 "2025-12-18T17:06:34Z")
I really wanted to like Gemini 3 Flash. I’ve been triple prompting between Grok 4.1, GPT 5.2 and Gemini 3 Flash Reasoning every since it came out. And Gemini 3 Flash hallucinates too much to be usable. I finally stopped …