LOADING DATASHEET
GROUNDED IN REAL HUMAN CONVERSATIONSYOUR BENCH
Claude Sonnet 4.6
by Anthropic
CLOSEDHOSTEDCODING ASSISTANTWRITING HELPLONG DOCUMENTSPROBLEM SOLVING
RANKED #212 OF 228 ASSISTANTS · OVERALL #325 OF 6,618 · VIBE SCORE 5.1 · 5,628 VOICES
A capable AI assistant built for writing, coding, and handling long texts.
HAPPY6.4
77 mentions
SAD8.6
95 mentions
VIBE TRAJECTORY
26 weeks · 8,099 voices
QUIETLY DEGRADING
Recent vibe and chatter are both sliding.
HOW TO USE IT
Closed model. Anthropic hosts it, so you use it through an app or an API.
PREFER OPEN SOURCE? THE CROWD HAS PICKS
Claude Sonnet 4.6 is closed, so you rent it. These open-weight models can genuinely stand in for it. Download once, own forever.
Llama 4 MaverickWHAT THE INTERNET SAYS
Users love Claude Sonnet 4.6 for coding help, natural writing, and working with long documents. However, people complain about high costs, frequent outages, occasional made-up facts, and overly strict refusals.
4294 POSTS · 1334 COMMENTS · REDDIT · HACKER NEWS · LEMMY · GITHUB · X · BLUESKY · DEV.TO · STACK OVERFLOW · DEV FORUMS · BLOG
PEOPLE LOVE
- Popular for coding help (26 mentions)
- Good at writing and wording things (8 mentions)
- Keeps up in long chats and big documents (3 mentions)
PEOPLE COMPLAIN ABOUT
- Outages, errors, or limits get in the way (32 mentions)
- People find it expensive (24 mentions)
- Sometimes makes things up (22 mentions)
- Refuses or lectures more than people like (15 mentions)
TOPIC BY TOPIC
- PriceMIXED FEELINGS
29 thumbs up · 24 thumbs down
- Working when you need itA SORE SPOT
7 thumbs up · 32 thumbs down
- Help with codeMOSTLY POSITIVE
26 thumbs up · 11 thumbs down
- Getting facts rightA SORE SPOT
8 thumbs up · 22 thumbs down
- Saying no too muchA SORE SPOT
2 thumbs up · 15 thumbs down
- WritingPEOPLE LOVE THIS
8 thumbs up · 2 thumbs down
- SpeedA SORE SPOT
1 thumbs up · 5 thumbs down
- Long conversationsPEOPLE LOVE THIS
3 thumbs up · 0 thumbs down
LOOKING FOR AN ALTERNATIVE?
Checked picks first: tags say why you would switch, stars say how fully each one stands in for Claude Sonnet 4.6.
Grok 4.6
GPT 5.6 Sol
GLM 5.3> SOURCES (12)
- REDDIT · 2026-06-18 · 12146 UPVOTESbuilt a factchecker that catches politicians lying in real time
built a factchecker that catches politicians lying in real time
- REDDIT · 2026-02-28 · 7408 UPVOTESCancel your Chatgpt subscriptions and pick up a Claude subscription.
Cancel your Chatgpt subscriptions and pick up a Claude subscription.
- REDDIT · 2026-04-07 · 4061 UPVOTESAnthropic's new model, Claude Mythos, is so powerful that it is not releasing it to the public.
Anthropic's new model, Claude Mythos, is so powerful that it is not releasing it to the public.
- REDDIT · 2026-02-28 · 2801 UPVOTESClaude is now 1st in the App Store
Claude is now 1st in the App Store
- REDDIT · 2026-02-23 · 2549 UPVOTESAnthropic is accusing DeepSeek, Moonshot AI (Kimi) and MiniMax of setting up more than 24,000 fraudulent Claude accounts, and distilling…
Anthropic is accusing DeepSeek, Moonshot AI (Kimi) and MiniMax of setting up more than 24,000 fraudulent Claude accounts, and distilling training information from 16 million exchanges.
- REDDIT · 2026-06-26 · 2443 UPVOTESI used Claude to fix my biggest frustration with PDFs
I used Claude to fix my biggest frustration with PDFs
- REDDIT · 2026-05-03 · 2118 UPVOTESRichard Dawkins spent 3 days with Claude and named her "Claudia." what he concluded after is hard to defend.
Richard Dawkins spent 3 days with Claude and named her "Claudia." what he concluded after is hard to defend.
- REDDIT · 2026-03-03 · 1637 UPVOTES90% of the world’s programmers when Claude goes down:
90% of the world’s programmers when Claude goes down:
- REDDIT · 2026-06-19 · 1227 UPVOTESAnyone prefer Claude over Gaming
Anyone prefer Claude over Gaming
- REDDIT · 2026-04-07 · 1123 UPVOTESClaude Mythos Was Told to Escape Sandbox in Testing — Succeeded, Then Unprompted Posted Exploit Details Online + Emailed Researcher While…
Claude Mythos Was Told to Escape Sandbox in Testing — Succeeded, Then Unprompted Posted Exploit Details Online + Emailed Researcher While He Was Eating a Sandwich in the Park
- REDDIT · 2026-03-08 · 1103 UPVOTESThe Washington Post: Claude Used To Target 1,000 Strikes In Iran
The Washington Post: Claude Used To Target 1,000 Strikes In Iran
- REDDIT · 2026-04-15 · 1081 UPVOTES🚨 RED ALERT: Tennessee is about to make building chatbots a Class A felony (15-25 years in prison). This is not a drill.
🚨 RED ALERT: Tennessee is about to make building chatbots a Class A felony (15-25 years in prison). This is not a drill.