X Feed Intel beta

individual tinkerer enterprises
Weekly Prep Posts Topics History
670 relevant 263 topics 1841 posts $1.088 cost fetched 02-23T21:39
670
Relevant
263
Topics
1841
Total Posts
$1.088
Cost This Week
$1.088
Total Cost
2026-02-23T21:39
Last Fetch
← Back to Topics
Frontier Models

Anthropic Claude Opus 4.6 release

Release and capabilities of Anthropic's Claude Opus 4.6 model, including performance improvements and user experiences.

9 posts · First seen 2026-02-23 · Last activity 2026-02-23
TimeAuthorPost
2026-02-23T21:10 @DanielleFong RT @TheZvi: I don't get it. Did you think Claude couldn't do that? Is this NEWS to anyone? Is the EMH that false?
2026-02-23T20:57 @TheZvi I don't get it. Did you think Claude couldn't do that? Is this NEWS to anyone? Is the EMH that false? https://t.co/hH7Lv50m1R
2026-02-23T20:28 @hive_echo RT @scaling01: Opus went from 20th place to 2nd place on undergraduate mathematics problems within 3 months (if you exclude all the Pro an…
2026-02-23T19:28 @scaling01 Opus went from 20th place to 2nd place on undergraduate mathematics problems within 3 months (if you exclude all the Pro and DeepThink models) https://t.co/x575xBI37G
2026-02-23T19:23 @Orwelian84 in many ways Anthropic is not to different from America-- they do all the wrong things before finally coming around to doing it right. early claude 2.0 with their bs safety baked to 11 was a trash tier model, they pivoted, and now in many ways claude is less censored than OAI -- i expect this distillation anti-china hawkishness will be a similar phase that they will pass through before settling on a more stable profitable configuration ↩ reply parent
2026-02-23T19:13 @DanielleFong this model is totally claude-strapped
2026-02-23T18:57 @corbtt Codex the model is great and the app seems quite well done. codex-5.3-xhigh gets essentially all of my coding usage at this point. That said, I still use Opus 4.6 daily in two workflows: 1. Much better browser automation 2. Better at the first draft of planning a major feature https://t.co/4y2rJnbKzn
2026-02-23T18:31 @Teknium @AnthropicAI https://t.co/UiYI1iwHwo ↩ reply parent
2026-02-23T17:33 @teortaxesTex Retranslated with Opus 4.6 high effort! enjoy, my favourite rant. Do you figure it aged well https://t.co/gj6JbJRkRU ↩ reply parent
@DanielleFong 2026-02-23T21:10
RT @TheZvi: I don't get it. Did you think Claude couldn't do that? Is this NEWS to anyone? Is the EMH that false?
@TheZvi 2026-02-23T20:57
I don't get it. Did you think Claude couldn't do that? Is this NEWS to anyone? Is the EMH that false? https://t.co/hH7Lv50m1R
@hive_echo 2026-02-23T20:28
RT @scaling01: Opus went from 20th place to 2nd place on undergraduate mathematics problems within 3 months (if you exclude all the Pro an…
@scaling01 2026-02-23T19:28
Opus went from 20th place to 2nd place on undergraduate mathematics problems within 3 months (if you exclude all the Pro and DeepThink models) https://t.co/x575xBI37G
@Orwelian84 2026-02-23T19:23
in many ways Anthropic is not to different from America-- they do all the wrong things before finally coming around to doing it right. early claude 2.0 with their bs safety baked to 11 was a trash tier model, they pivoted, and now in many ways claude is less censored than OAI -- i expect this distillation anti-china hawkishness will be a similar phase that they will pass through before settling ...
↩ reply parent
@DanielleFong 2026-02-23T19:13
this model is totally claude-strapped
@corbtt 2026-02-23T18:57
Codex the model is great and the app seems quite well done. codex-5.3-xhigh gets essentially all of my coding usage at this point. That said, I still use Opus 4.6 daily in two workflows: 1. Much better browser automation 2. Better at the first draft of planning a major feature https://t.co/4y2rJnbKzn
@Teknium 2026-02-23T18:31
@AnthropicAI https://t.co/UiYI1iwHwo
↩ reply parent
@teortaxesTex 2026-02-23T17:33
Retranslated with Opus 4.6 high effort! enjoy, my favourite rant. Do you figure it aged well https://t.co/gj6JbJRkRU
↩ reply parent

Markdown Export

Loading...