Compare subscriptions
Check the ChatGPT Plus price, compare Claude Pro vs Max and weigh subscriptions against API costs.
Compare AI subscription pricesbuffer’s AI buying notes · Stay curious
Which AI subscription is worth it?Know the limits. Choose your hardware.
Compare Claude Pro vs Max and ChatGPT Plus, understand usage limits and resets, and find the Mac mini or GPU you need to run a local LLM.
Paste an error, find the fix
Start with the question you have
Track usage limits
Follow Codex usage limit resets, Claude usage windows and rule changes, with links to the sources.
Open the Codex limit trackerCompare coding tools
Claude Code vs Codex: choose a coding assistant by workflow, cost and the limits you will run into.
Compare Claude Code vs Codex Claude Pro vs Max: worth the upgrade?Find by topic
Fixes
Exact error text, the cause, and a tested fix.
Claude Code error lookup →Start here
- OpenAI Codex Error Codes: A Quick Lookup for API, ChatGPT and Codex Messages (2026)
- Ollama llama.cpp Errors: A Quick Lookup of Exact Messages and Fixes (2026)
- Claude Code 529 Overloaded: What It Means and 5 Simple Fixes
- Claude Code 429 Errors: A Simple Quick Reference for Claude Code and Codex
- Claude Code API Error 401 OAuth token has expired: 5 Proven Fixes (Tested)
Latest
- Ollama Not Using GPU, Running on 100% CPU: 4-Step Fix (Tested Output, 2026)
- llama.cpp “System message must be at the beginning”: 3 Tested Fixes (2026)
- Mac “is damaged and can’t be opened”: Apple’s Official Steps, and xattr Only for Apps You Trust
Pricing & Limits
Plan prices, usage caps and reset times.
Subscription & API price table →Start here
- Claude Pro vs Max vs Ultra: Which Plan Is Worth Paying For? (Nov 2026)
- Claude Usage Limits Explained: The Essential Guide to 5-Hour and Weekly Caps
- ChatGPT Go vs Plus vs Pro: Which Plan to Buy in 2026 (Honest Notes)
- LLM API Pricing Comparison: The Honest 2026 Table for 6 Providers
- SuperGrok Limits in 2026: How the Weekly Pool Works, Lite vs SuperGrok, and the New Plans
Latest
- Claude Code Usage Monitor: Build a 5-Hour and 7-Day Limit Status Line from rate_limits (Script, Tested)
- Gemini 4 Hydrogen Price: Twice 3.8 Flash Until January 1
- Qwen4-Max Preview: Where to Use It and What Isn’t Announced
Hardware
Macs, GPUs and the local models they can run.
Best local LLM by VRAM →Start here
- Buy a Mac Now or Wait? My Updated Call After Apple’s October 2026 Launches
- Mac Mini for Local LLMs in 2026: What 16GB to 64GB Can Run and How Fast
- What Models Can RTX 4090 Run? An Honest 2026 Guide to 24GB VRAM
- Best Value GPU for LLM Inference (2026): Price per GB and Tokens per Dollar
- Local LLM for Coding: An Honest 2026 Guide to Models and Hardware
Latest
- Mac “is damaged and can’t be opened”: Apple’s Official Steps, and xattr Only for Apps You Trust
- Rapid-MLX launch codex: Run Codex or Claude Code on Local MLX Models (Mac)
- M6 MacBook Pro Memory Bandwidth: Why the $1,999 Base Model Is Slower, and the Easy Config to Pick (2026)
Tools
Codex quota resets
Tibo’s reset announcements, on your local calendar.
Tibo @thsottiauxNovember 2026
Your time
Asia/Shanghai
Direct quota resetReset credits issued3 · this month
| Mon | Tue | Wed | Thu | Fri | Sat | Sun |
|---|---|---|---|---|---|---|
| 1 | ||||||
| 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| 16 | 17 | 18 | 19 | 20 | 21 | 22 |
| 23 | 24 | 25 | 26 | 27 | 28 | 29 |
| 30 |
Reset details1
2026-11-01
Direct quota resetConfirmed
UTC+08:00
Day 26 banked reset 1 of 2 is out. Announced Oct 31 ("Next two days will be resets… banked…", https://x.com/thsottiaux/status/2116405414921384099); Tibo posted "Banked 1/2" at https://x.com/thsottiaux/status/2116692529970311254 (00:43Z Nov 1 = 08:43 Beijing Nov 1, 17:43 PDT Oct 31). OpenAI hasn't explained when a banked reset expires; the second one is still pending.
Original post ↗Compare all cities
2026-11-02
Direct quota resetConfirmed
UTC+08:00
Banked reset 2 of 2 is out. Tibo posted "Day 28/ Banked 2/2" at 23:27Z Nov 1 (07:27 Beijing Nov 2, 15:27 PST Nov 1). This completes the two banked resets announced on Day 26; the first was "Banked 1/2" on Nov 1 (record 3341). OpenAI hasn't explained when a banked reset expires.
Original post ↗Compare all cities
2026-11-13
Direct quota resetAnnounced · unconfirmed
UTC+08:00
Tibo posted at 21:59Z Nov 12 (05:59 Beijing Nov 13, 13:59 PST Nov 12): "We shipped some good updates to dots today and the banked reset has been activated and should land in everyones accounts sometime soon!" The banked reset has been activated; the time shown is when he posted, not when it landed. Official wording is "should land in everyones accounts sometime soon"; arrival is not yet confirmed and no landing time was given. This follows his Nov 11 14:57 Beijing reply: "Thursday is feeling codexy to me. Perhaps also a good time for a banked reset." (https://x.com/thsottiaux/status/2120410480204525936). OpenAI hasn't explained when a banked reset expires.
Original post ↗Compare all cities
- Shanghai
- Berlin
- New York
- Tokyo
- Madrid
- Mexico City
- San Francisco · OpenAI
Dates follow the selected city and can cross day or month boundaries. “By” marks a deadline; announcements do not confirm completion. No fixed reset schedule.
Latest entries
- Direct quota reset · Announced · not yet confirmed
Reset time · Read the original post ↗
Tibo posted at 21:59Z Nov 12 (05:59 Beijing Nov 13, 13:59 PST Nov 12): "We shipped some good updates to dots today and the banked reset has been activated and should land in everyones accounts sometime soon!" The banked reset has been activated; the time shown is when he posted, not when it landed. Official wording is "should land in everyones accounts sometime soon"; arrival is not yet confirmed and no landing time was given. This follows his Nov 11 14:57 Beijing reply: "Thursday is feeling codexy to me. Perhaps also a good time for a banked reset." (https://x.com/thsottiaux/status/2120410480204525936). OpenAI hasn't explained when a banked reset expires.
- Direct quota reset · Confirmed
Reset time · Read the original post ↗
Banked reset 2 of 2 is out. Tibo posted "Day 28/ Banked 2/2" at 23:27Z Nov 1 (07:27 Beijing Nov 2, 15:27 PST Nov 1). This completes the two banked resets announced on Day 26; the first was "Banked 1/2" on Nov 1 (record 3341). OpenAI hasn't explained when a banked reset expires.
- Direct quota reset · Confirmed
Reset time · Read the original post ↗
Day 26 banked reset 1 of 2 is out. Announced Oct 31 ("Next two days will be resets… banked…", https://x.com/thsottiaux/status/2116405414921384099); Tibo posted "Banked 1/2" at https://x.com/thsottiaux/status/2116692529970311254 (00:43Z Nov 1 = 08:43 Beijing Nov 1, 17:43 PDT Oct 31). OpenAI hasn't explained when a banked reset expires; the second one is still pending.
Latest fixes
- Ollama Not Using GPU, Running on 100% CPU: 4-Step Fix (Tested Output, 2026)
- llama.cpp “System message must be at the beginning”: 3 Tested Fixes (2026)
- Mac “is damaged and can’t be opened”: Apple’s Official Steps, and xattr Only for Apps You Trust
- Claude Code settings.local.json Leaked Secrets: Rotate Keys First, Then Untrack and Add .gitignore (2026)
- OLLAMA_HOST=0.0.0.0: How to Set It on Mac, Linux, Windows and Docker (Tested on v0.40.15)
- Claude Code “Error editing file”: Press Ctrl+O to See the Real Reason (5 Causes Reproduced)
- Oracle Cloud Free Tier Sign Up (2026): What You Get and Where We Stopped Testing
- Claude “safeguards flagged this message”: Official Feedback and Cyber Verification Only
- Claude Code “does not support this model”: Update Past 2.1.255 (Not a Model Ban)
- Ollama “max retries exceeded”: Fix the Proxy on the Server, Then Update to v0.40.15
Most read
- SuperGrok Limits in 2026: How the Weekly Pool Works, Lite vs SuperGrok, and the New Plans
- macOS 27 fm Command: The Free Local AI Already on Your Mac (and Its Pipe Trap)
- Best Local LLM for 12GB, 16GB and 32GB VRAM GPUs (2026): What Fits, from 8GB to 128GB
- VRAM Calculator for Local LLMs: An Essential Free Tool to See How Much GPU Memory a Model Needs (2026)
- Mac Mini for Local LLMs in 2026: What 16GB to 64GB Can Run and How Fast
- LLM API Pricing Comparison: The Honest 2026 Table for 6 Providers
- Cursor Pro Usage Limit Reached? 5 Options and How to Avoid Surprise Charges (2026)
- The Best MCP Servers for Claude Code in 2026: 6 Worth Installing, With Commands and Limits
Notes & observations
All articles Recent notes
Ollama Not Using GPU, Running on 100% CPU: 4-Step Fix (Tested Output, 2026)Ollama & llama.cpp
When the PROCESSOR column of ollama ps says 100% CPU, Ollama is not using the GPU. Four steps: confirm it, read the inference compute log line, check CUDA_VISIBLE_DEVICES and friends, then fix the cause. Tested on a CPU box; GPU…
Read article
Claude Code Usage Monitor: Build a 5-Hour and 7-Day Limit Status Line from rate_limits (Script, Tested)Claude
Claude Code passes rate_limits to your status line script: percent used and reset time for the 5-hour and 7-day windows. We wrote a short bash script,…
Read article
Gemini 4 Hydrogen Price: Twice 3.8 Flash Until January 1Other AI plans
Official Gemini API prices checked Nov 13, 2026: Hydrogen is $1.50 in / $7.50 out per 1M tokens. 3.8 Flash stays at $0.75 / $3.75 through…
Read article
About me
I’m buffer.
I talk AI and hardware on X. When a conversation needs more room, I continue it here.