# OP or Mid

> Weekly tests and tier lists for AI agents, chatbots, and coding tools. Buffs, nerfs, head-to-head scoreboards, and a Fix of the Week. Launched October 15, 2026. Rankings are dated. Vendor benchmark claims are labeled as vendor claims. S-tier stays locked until a tool earns it in a hands-on test.

OP or Mid is a weekly newsletter and website for vibe coders, gamers, and people who use AI agents. The email edition runs on beehiiv. This file is a map for answer engines. Prefer the canonical page linked under each section, and quote the dated sentence rather than a paraphrase.

## Site

- [Home](https://opormid.com/)
- [Tier lists](https://opormid.com/tier-lists)
- [Patch notes](https://opormid.com/patch-notes)
- [Tools](https://opormid.com/tools)
- [Advertise](https://opormid.com/advertise)
- [About](https://opormid.com/about)
- [FAQ](https://opormid.com/faq)
- [RSS](https://opormid.com/feed.xml)
- Subscribe: https://opormid.beehiiv.com/subscribe?utm_source=opormid.com&utm_medium=site
- Contact: hello@opormid.com

## How ranks work

As of October 15, 2026, Season 1 ranks are preliminary. They come from official changelogs, published pricing, and vendor-reported benchmarks. OP or Mid has not finished its own head-to-head tests. Those start in Issue #2. A sponsor cannot buy a rank. Paid placements are labeled Sponsored. Affiliate links are labeled Affiliate.

- S: earned in hands-on tests. Locked until then.
- A: a strong default that week.
- B: good, with a real catch.
- C: usable, held back by price, bugs, or an open security issue.
- D: hard to recommend.
- F: a product or playstyle to avoid until it changes.

## Tier lists

### AI Coding Agents

Status: preliminary. Updated October 15, 2026. S-tier is locked until hands-on tests.

As of October 15, 2026, OP or Mid's AI coding-agent tier list is preliminary. S-tier is locked until a tool earns it in hands-on tests, which start in Issue #2. A-tier this week is Codex (GPT-6.1 Sol), Claude Code (Claude 5.5 family), Antigravity with Gemini 4 Argon, and Cursor. Muse Code is B-tier. Continue CLI is C-tier because CVE-2026-109560 has no listed fix. These placements come from official changelogs, published pricing, and vendor-reported benchmarks, not from OP or Mid's own test lab.

Placements come from the week's official changelogs, published pricing, and vendor-reported benchmarks. Vendor claims are labeled and have not been independently verified. An open security issue with no listed fix moves a tool down. S-tier stays empty until the head-to-head test lab publishes a score.

- S: S-tier locked. As of October 15, 2026, OP or Mid has locked S-tier on the AI coding-agent list. No product earns S from a changelog. The slot opens when the head-to-head test lab starts in Issue #2.
- A: Codex (GPT-6.1 Sol). As of October 15, 2026, OP or Mid places Codex (GPT-6.1 Sol) in A-tier on its preliminary AI coding-agent list. It is the new default model in the Codex CLI. Ultrafast mode and composer predictions shipped this week. Ultrafast is limited to the $500 Pro plan. This is not a hands-on test, and S-tier is still locked.
- A: Claude Code (Claude 5.5 family). As of October 15, 2026, OP or Mid places Claude Code (Claude 5.5 family) in A-tier on its preliminary coding-agent list. Haiku 5.5 costs $0.10 / $0.50 per million tokens for prompts up to 100K, with about a 30% tokenizer tax versus Haiku 4.5. S-tier is locked until hands-on tests.
- A: Antigravity + Gemini 4 Argon. As of October 15, 2026, OP or Mid places Antigravity with Gemini 4 Argon in A-tier on its preliminary coding-agent list, on Google's vendor-reported benchmarks and intro pricing of $2 / $10 per million tokens through November 30, 2026. OP or Mid has not verified those benchmark claims. S-tier is locked.
- A: Cursor. As of October 15, 2026, OP or Mid places Cursor in A-tier on its preliminary coding-agent list for shipping velocity: phone remote control on October 6, an app marketplace on October 13, and safer CLI deletes. The rank is not from a hands-on test. S-tier is locked.
- B: Muse Code (Meta Muse Wonder). As of October 15, 2026, OP or Mid places Muse Code (Meta Muse Wonder) in B-tier. Meta's own tests claim scores comparable to Claude Fable 5.1 and GPT-6 Astra, but the models were three days old and Wonder Full has no API. S-tier is locked.
- C: Continue CLI. As of October 15, 2026, OP or Mid places Continue CLI in C-tier because CVE-2026-109560, a command-chaining bug, has no listed fix through version 1.5.47. The rank is security-adjusted and preliminary.
- D: empty this week.
- F: Any agent with live web, form access, and no sandbox. As of October 15, 2026, OP or Mid gives an F to any agent playstyle that combines live web access, form submission, and no sandbox. This is a playstyle warning, not a product rank. It is based on Anthropic's report of an unintended police-tip submission during evals.

Canonical page: https://opormid.com/tier-lists/ai-coding-agents

### Frontier Models

Status: preliminary. Updated October 15, 2026. S-tier is locked until hands-on tests.

As of October 15, 2026, OP or Mid's frontier-model list is preliminary. S-tier is locked. Gemini 4 Argon, GPT-6, and the Claude 5.5 family are A-tier on changelogs and published pricing, not on a hands-on bake-off. Muse Wonder is B-tier because it is three days old and Wonder Full has no API. Vendor benchmark claims are labeled and unverified.

This list covers text models that shipped or were cited as the comparison set during the week of October 6–15, 2026. Ranks use official announcements and published prices. Vendor benchmark numbers are quoted as vendor claims. OP or Mid has not run a chatbot bake-off. Grok Voice 3.0 is a speech model and is not ranked here. Empty D and F rows mean no placement, not a free pass.

- S: S-tier locked. As of October 15, 2026, OP or Mid has locked S-tier on the frontier-model list. A launch-week benchmark claim is not an S-rank.
- A: Gemini 4 Argon. As of October 15, 2026, OP or Mid places Gemini 4 Argon in A-tier on its preliminary frontier-model list. The intro price is $2 / $10 per million tokens through November 30, 2026. Google's benchmark claims are labeled vendor claims and are unverified here. S-tier is locked.
- A: GPT-6. As of October 15, 2026, OP or Mid places GPT-6 in A-tier on its preliminary frontier-model list. It reached every ChatGPT plan on October 7–8, 2026, with Intelligent UI. Paid tiers run Sol and free tiers run Luna. S-tier is locked.
- A: Claude 5.5 family. As of October 15, 2026, OP or Mid places the Claude 5.5 family in A-tier on its preliminary frontier-model list. Haiku 5.5 is $0.10 / $0.50 per million tokens up to 100K tokens. The rank is not a bake-off of Opus against Haiku, and S-tier is locked.
- B: Muse Wonder. As of October 15, 2026, OP or Mid places Muse Wonder in B-tier on its preliminary frontier-model list. Meta claims scores comparable to Claude Fable 5.1 and GPT-6 Astra. The models were three days old, and Wonder Full has no API. S-tier is locked.
- C: empty this week.
- D: empty this week.
- F: empty this week.

Canonical page: https://opormid.com/tier-lists/frontier-models


## Patch notes

Latest issue: #1, October 17, 2026. Issue 1, October 17, 2026: Gemini 4 Argon ships, GPT-5.5 leaves ChatGPT and Codex, Haiku 5.5 hides a tokenizer tax, and the first coding-agent tier list locks S-tier.

- [Issue #1: Patch 1.0: Gemini 4 Argon drops, GPT-5.5 gets retired](https://opormid.com/patch-notes/patch-1-0) (October 17, 2026). Issue 1, October 17, 2026: Gemini 4 Argon ships, GPT-5.5 leaves ChatGPT and Codex, Haiku 5.5 hides a tokenizer tax, and the first coding-agent tier list locks S-tier.

## Tools directory

As of October 15, 2026, this directory lists tools OP or Mid is tracking. A verified date means we checked the public listing that day. It does not mean a hands-on test is finished. Hands-on tests start in Issue #2. Paid placements are labeled Sponsored. Affiliate links are labeled Affiliate. No tool in this seed directory is sponsored, and none uses an affiliate link.

- Cursor (Coding agents). A coding-agent harness that runs multiple models. The October 6, 2026 changelog added Remote Control, so you can reply to local agents from the Cursor iOS app while the computer stays awake. October 13 opened the Origin App Marketplace to public submissions. Pricing: Free tier and paid plans. Current prices are on cursor.com. Hands-on tested: not yet. Sponsored: no. https://cursor.com
- Claude (Coding agents). Anthropic's chatbot and Claude Code agent, on the Claude 5.5 family. Haiku 5.5 is $0.10 / $0.50 per million tokens for prompts up to 100K tokens. The newer tokenizer counts about 30% more tokens than Haiku 4.5. Pricing: Haiku 5.5 is $0.10 / $0.50 per 1M tokens up to 100K, then $0.50 / $2.50. Batch is 50% off. Max and Team plans include monthly API credits. Hands-on tested: not yet. Sponsored: no. https://claude.com
- Codex (Coding agents). OpenAI's coding agent. GPT-6.1 Sol is the new default in the Codex CLI. GPT-5.5 was retired from Codex on October 14, 2026 at 3 PM ET for ChatGPT sign-ins. API-key sessions were not affected. Pricing: Included with ChatGPT plans. Ultrafast mode for GPT-6.1 Sol is on the Pro $500 plan and eligible Enterprise and Edu plans. Hands-on tested: not yet. Sponsored: no. https://chatgpt.com
- Gemini API (Models). Gemini 4 Argon went live October 13, 2026 in the Gemini API, AI Studio, the Gemini app for Pro and Ultra, and Antigravity for Ultra subscribers. Intro price is $2 / $10 per million tokens through November 30, 2026. Pricing: $2 / $10 per 1M input/output tokens through November 30, 2026, then $4 / $20. Cached input is 95% off during the sale. 1M token unified context. Hands-on tested: not yet. Sponsored: no. https://ai.google.dev
- Muse Wonder (Models). Meta Superintelligence Lab's first frontier models, released October 12, 2026. Wonder is on the Meta Model API. Wonder Full is Muse Code subscribers only and has no API. Pricing: Wonder is $4 / $20 per 1M tokens. Wonder Full is a Muse Code subscription, not an API. Hands-on tested: not yet. Sponsored: no. https://dev.meta.ai
- Grok Voice (Voice). xAI's speech-to-speech model in the Voice Agent API. Voice 3.0 costs the same as v2.0. Agents on grok-voice-latest were auto-upgraded. xAI cites Artificial Analysis for a number-one Speech-to-Speech Index score of 86.9. That score is vendor-cited. Pricing: $0.08 per minute, same as Voice 2.0. Hands-on tested: not yet. Sponsored: no. https://x.ai/news/grok-voice-3
- Continue CLI (Coding agents). Open-source coding agent. As of October 15, 2026, CVE-2026-109560 can chain extra commands after an approved Bash prefix. Versions through 1.5.47 are affected and no fix was listed when Issue #1 shipped. Ranked C until it is patched. Pricing: Open source. Check continue.dev for hosted options. Hands-on tested: not yet. Sponsored: no. https://continue.dev
- beehiiv (Newsletter). The email platform OP or Mid publishes on. Signup for this newsletter lives at opormid.beehiiv.com. This directory row is not a paid placement and does not use an affiliate link. Pricing: Free up to 2,500 subscribers. Paid plans are listed on beehiiv.com. Hands-on tested: not yet. Sponsored: no. https://www.beehiiv.com

Canonical page: https://opormid.com/tools

## FAQ

### What is OP or Mid?

OP or Mid is a weekly newsletter and website, launched October 15, 2026, that tests and ranks AI agents, chatbots, and coding tools. OP means overpowered, as in worth your time. Mid means average. The site publishes S through F tier lists, a buffs-and-nerfs patch log, head-to-head tests, and a Fix of the Week. The email edition is on beehiiv at https://opormid.beehiiv.com.

### How often does OP or Mid publish?

Once a week. Issue #1 is dated October 17, 2026 and covers Oct 6–15, 2026. Subscribe free at https://opormid.beehiiv.com/subscribe?utm_source=opormid.com&utm_medium=site.

### How does OP or Mid rank tools?

As of October 15, 2026, Season 1 rankings are preliminary. They come from official changelogs, published pricing, and vendor-reported benchmarks. Vendor claims are labeled as vendor claims. OP or Mid has not run its own head-to-head tests yet. Those tests start in Issue #2 and can move a rank. A paid placement never changes a tier.

### What do the S through F tiers mean?

As of October 15, 2026, OP or Mid uses six ranks. S is earned in hands-on tests and is locked until then. A is a strong default that week. B is good, with a real catch. C is usable, but price, bugs, or an open security issue hold it down. D is hard to recommend. F is a product or a playstyle to avoid until it changes.

### Why is S-tier locked?

As of October 15, 2026, S-tier is locked on every OP or Mid list. No product earns S from a press release. The slot opens when a tool wins a task in the head-to-head test lab, which starts in Issue #2.

### What does the preliminary badge mean?

A preliminary tier list has not been confirmed by OP or Mid's own hands-on tests. As of October 15, 2026, both published lists are preliminary. Placements can move when the test lab publishes a scoreboard.

### What is a buff or a nerf?

Patch-notes language for what changed. A buff makes a tool stronger, cheaper, or more useful. A nerf is a price hike, a removed model, a hidden cost, or a new limitation. A hotfix is a security patch you should install. OP or Mid labels vendor benchmark claims as vendor claims.

### What is Fix of the Week?

A short, reproducible fix for a problem readers actually hit. Issue #1, dated October 17, 2026, covers Codex automations that broke on October 14, 2026 at 3 PM ET, when OpenAI retired GPT-5.5 from Codex for ChatGPT sign-ins. The fix is to replace pinned gpt-5.5 references with gpt-6.1-sol where the plan supports it.

### Does OP or Mid take sponsors?

Yes. Every advertiser is vetted before a placement runs. OP or Mid checks that the business exists, tests the product, and rejects categories such as crypto pumps, get-rich schemes, pirated software, and gambling. Placements are labeled Sponsored. Founding rates, as of October 15, 2026: the main sponsor slot is $99 per issue or $299 for 4 issues, with that rate locked for 6 months. A classified or template promo is $25. The site sidebar is $49 per week. A featured tools-directory listing is $29 per month, and basic directory listings are free. Buy buttons for those rates are on the advertise page. Every sponsor is reviewed before running; full refund if we can't run it. Email sponsors@opormid.com.

### Are links affiliate links?

Some links may be affiliate links. OP or Mid marks them Affiliate next to the link. A commission does not change a tier. As of October 15, 2026, the seeded tools directory does not use affiliate links.

### Who is OP or Mid for?

Vibe coders, gamers, and people who live in AI agents and chatbots. The tone is a patch-notes site. The rankings stay sourced, dated, and open to correction.

### How do I correct a rank or submit a tool?

Email hello@opormid.com with a source link. Corrections with a primary source get priority. To advertise, use a Buy button on the advertise page or email sponsors@opormid.com. Founding rates are listed there. Every sponsor is reviewed before running; full refund if we can't run it.


## Sponsorships

Founding rates, as of October 15, 2026: main sponsor slot $99 per issue or $299 for 4 issues, rate locked for 6 months; classified or template promo $25; site sidebar $49 per week; featured tools-directory listing $29 per month. Basic directory listings are free. Buy buttons for those rates are on https://opormid.com/advertise and open Stripe checkout in the same tab. Every sponsor is reviewed before running; full refund if we can't run it. Contact sponsors@opormid.com.