Matching posts

Matching terms are highlighted

Which metric actually matters when comparing LLMs like DeepSeek V4.1

The author argues token prices, benchmark scores and speed don't capture the real cost of a verified working deliverable, using a bridge-steel analogy and calling for cost-per-verified-task comparisons.

deepseek-v4.1@TugserOkur　Filed in No.020

21 hours agoCopy original

POST

DeepSeek V4.1 Flash open weights and MIT license prompt debate on closed models

Using a $2 Blender render as a hook, the post notes V4.1 Flash ships open weights under MIT at the same low price with better reasoning, coding and agents, and asks whether open weights are now the default.

deepseek-v4.1@0xDrexo　Filed in No.020

16 hours agoCopy original

POST

Where Chinese open models fall short: long context and agent stability

A discussion thread concludes Chinese open models struggle with long-context needle-in-haystack and agent tool-call stability, though DeepSeek V4.1 holds up on single-file algorithms, regex and SQL.

deepseek-v4.1@ShokalerP　Filed in No.020

14 hours agoCopy original

POST

Claude Duet's proactive background work raises trust question

The post notes that Claude Duet wants to work proactively in the background and asks whether users would trust an AI that starts tasks before being asked.

claude@emaann28　Filed in No.012

1 day agoCopy original

Claude Duet waitlist: how many have been let in

A post asks how many people are still on the Claude Duet waitlist or haven't applied, to gauge the actual let-in rate.

claude@ConnorTalksAI　Filed in No.012

1 day agoCopy original

POST

Grok Bot bundled into X Pass makes cancellation unlikely

A user argues they would keep paying for X Pass even if Claude Duet outclasses Grok Bot, since Premium+, Cursor Pro and creator payouts are bundled in.

grok@aGim_asf　Filed in No.012

16 hours agoCopy original

POST

Grok Bot's moat: X integration over coding ability

The author argues Grok Bot wins on X's finance and news integration rather than coding, and can thrive like Muse by serving its own ecosystem's users.

personal-agent@Web3Feng　Filed in No.012

20 hours agoCopy original

Grok Bot communities: who has joined so far

An image post asking who has already joined the Grok Bot communities, pointing readers toward where actual users are gathering.

personal-agent@HeyYoGB　Filed in No.012

1 day agoCopy original

Could Grok Bot become the household-name AI assistant?

The author argues many people who still don't use AI may onboard through @bot rather than GPT or Claude, and asks whether Grok Bot becomes the household name.

personal-agent@zhodonx　Filed in No.012

21 hours agoCopy original

POST

Reply notes Claude bans Chinese accounts, not French ones

A reply argues Claude only bans Chinese accounts and not French ones, questioning why French researchers don't distill from OpenAI instead.

claude@mitechen　Filed in No.011

1 day agoCopy original

Showing10of60posts

10 per batch

Keep scrolling to reveal posts 11–20

Showing the first 60 matches. Narrow the query with a model name or tag to find more specific results.