Raphi-2Code retweeted
In AA-PersonalityBench, GPT-6 and Grok models are the most Direct and most likely to use an Answer First format of all models measured
The Direct trait assesses whether models get to the point quickly without unnecessary filler; Answer First measures whether models lead with the additional context, reasoning or caveats.
➤ Most Direct: GPT-6 Sol (100), Grok 4.6 (98) and Grok 4.7 (95)
➤ Most likely to use an Answer First form: Grok 4.5 (100), GPT-6 Sol (98) and GPT-6 Luna (95)
The Direct trait assesses whether models get to the point quickly without unnecessary filler; Answer First measures whether models lead with the additional context, reasoning or caveats.
➤ Most Direct: GPT-6 Sol (100), Grok 4.6 (98) and Grok 4.7 (95)
➤ Most likely to use an Answer First form: Grok 4.5 (100), GPT-6 Sol (98) and GPT-6 Luna (95)
31
17
365
38,548
Raphi-2Code retweeted
The first signs of "GPT Image 2.6" have started showing up, and OpenAI is getting ready to launch its next generation of image models.
I believe this could be one of the next releases we'll see, alongside new frontier models that I also expect to arrive in the coming days.
I believe this could be one of the next releases we'll see, alongside new frontier models that I also expect to arrive in the coming days.
22
10
295
29,916
Raphi-2Code retweeted
Exciting news: Gemini 4 Carbon (High) from @GoogleDeepMind just made a huge debut on Arena, landing #1 in Text Arena, #2 in Agent Arena, and #4 in Code Arena: WebDev. Back at the top, Gemini 4 Carbon delivers frontier intelligence and reshapes the Pareto frontier across three , Gemini 4 Carbon ranks #2 with a +16.2% net improvement across +21.9K real-world agentic sessions from our global community, surpassing top models GPT-6 Astra and Fable-5.1. At $1.80 median cost per task, Gemini 4 Carbon (High) costs 72% less than the #1 ranked Claude Opus 5.5 (Max).
Its standouts by key signals:
- #1 Steerability (+18.6%): can the model course-correct when you push back?
- #2 Praise vs Complaint (+37.9%): implicit sentiment in users' reactions
- #3 Confirmed Success (+18.7%): explicit “yes, that worked” from users
In Code Arena: WebDev, Carbon scored 1768 pts landing #4:
- Just 2 pts below #3 Claude Sonnet 5.5 (xHigh), at 1770 pts
- 89 pts above #11 Gemini 4 Argon (High), at 1679 pts
By category, it ranks #3 in Gaming and Simulations, and #5 in Brand & Marketing.
In Text Arena, Gemini 4 Carbon (High) is the new #1 with 1528 pts, now ahead of Gemini 4 Argon (High) at #2 with 1520 pts.
Congrats to the @GoogleDeepMind team on this huge milestone!
Its standouts by key signals:
- #1 Steerability (+18.6%): can the model course-correct when you push back?
- #2 Praise vs Complaint (+37.9%): implicit sentiment in users' reactions
- #3 Confirmed Success (+18.7%): explicit “yes, that worked” from users
In Code Arena: WebDev, Carbon scored 1768 pts landing #4:
- Just 2 pts below #3 Claude Sonnet 5.5 (xHigh), at 1770 pts
- 89 pts above #11 Gemini 4 Argon (High), at 1679 pts
By category, it ranks #3 in Gaming and Simulations, and #5 in Brand & Marketing.
In Text Arena, Gemini 4 Carbon (High) is the new #1 with 1528 pts, now ahead of Gemini 4 Argon (High) at #2 with 1520 pts.
Congrats to the @GoogleDeepMind team on this huge milestone!
Gemini 4 Carbon is our next frontier model.
It’s state-of-the-art across cybersecurity and long-horizon capabilities.
74
168
2,172
228,119
Gemini 4 Carbon SVG peacock
Google DeepMind announced Gemini 4 Carbon! It tops the Artificial Analysis Intelligence Index at 59.4 and costs marginally more per task than Argon. I’m wary of Google potentially benchmaxxing, but it also looks strong on Arena. Can’t wait to test it. Huge congrats to the team!
6
5
66
2,698
Google DeepMind announced Gemini 4 Carbon! It tops the Artificial Analysis Intelligence Index at 59.4 and costs marginally more per task than Argon. I’m wary of Google potentially benchmaxxing, but it also looks strong on Arena. Can’t wait to test it. Huge congrats to the team!
0
0
7
2,788
Raphi-2Code retweeted
🚨 BREAKING: DeepMind have announced Gemini 4 Carbon. Per these benchmarks, it’s the new best model in the world, and leads Artificial Analysis’ Intelligence Index v5
Introducing Gemini 4 Carbon, it's a big jump over Argon which we released just 2 weeks ago.
Gemini 4 Carbon is SOTA on cyber, long horizon tasks, ranks #1 on Artificial Analysis V5, significantly improves on usability in domains like coding, and is more aligned.
37
24
911
68,046
@Dan6i5 @dummerspast39 that doesn’t mean the score is not legit, he just created this using oai tools for visualization
1
0
0
61
On high effort, GPT-6.1 Sol is pretty much the best model on the Artificial Analysis Index, respectively Grok 4.7 the worst.
The AA-Intelligence Index has been updated to V5
Rankings before:
1. Claude Opus 5.5 (Max) - 58
2. Claude Sonnet 5.5 (Max) - 56
3. Muse Wonder Full (Max)- 56
4. Kimi K3.1 Pro - 55
5. Claude Fable 5.1 (Max) - 53
Rankings now:
1. Claude Opus 5.5 (Max) - 58
2. GPT-6-Astra (Max) x)- 55
4. Claude Sonnet 5.5 (Max) - 54
5. GPT-6.1-Terra (Max) - 51
2
2
147
18,868
@Angaisb_ For me, it doesn’t:
For me Grok 4.7 is better than Claude Haiku 5.5, not worse and GPT-6.1 Terra is worse than Gemini 4 Argon imo.
For me Grok 4.7 is better than Claude Haiku 5.5, not worse and GPT-6.1 Terra is worse than Gemini 4 Argon imo.
0
0
0
56
Artificial Analysis Index v5!!!!! 🥳
Announcing Artificial Analysis Intelligence Index v5, featuring 2 new benchmarks and 3 upgrades. For the first time, the majority of the Index is built on private datasets
The Artificial Analysis Intelligence Index is our composite measure of model capabilities across tasks sage today. v5 is our biggest upgrade since v4 launched in January. With v5, 60% of the Index is now based on evaluations with private test sets or held-out answers.
Alongside Intelligence Index v5, we’re launching the Artificial Analysis Progress Index v1, which tracks model capability over time on one continuous scale.
New to the Intelligence Index:
➤ FrontierCode 2.0: Cognition's newly-launched private benchmark of real-world coding tasks, graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench-Science: 70 agentic scientific research tasks, run in a terminal environment
Upgraded:
➤ HLE-Diamond (replacing Humanity's Last Exam): a refined 1,000-question version of Humanity's Last Exam, split evenly between reasoning and knowledge questions
➤ AA-Omniscience v1.1: An updated dataset with saturated and public set questions removed, new questions added, and ambiguous or erroneous questions revised
➤ CritPt v2: a private, 40-question hard subset of CritPt v2, revised by the CritPt authors and run with a code-execution tool
Staying in the Index: AA-Briefcase, AutomationBench-AA, GDP.pdf and Terminal-Bench 4.0
Removed from the Index: GDPval-AA v2.1, SciCode, and AA-LCR v1.1
Key results:
➤ Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol (max, 55): Anthropic and OpenAI hold the top five models. Gemini 4 Argon (high) is next at 50
➤ The open weights gap widens from 3 points to 12: Kimi K3.1 Pro (weights not yet released) is the top open weights model at 46. Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32 - 26 points behind the leader and below Claude Haiku 5.5 (max, 36)
➤ Cost per task varies by more than 13x among the leading models: of the 10 highest-scoring models shown, GPT-6.1 Terra (max) is the cheapest at $0.96 per Intelligence Index task and Claude Fable 5.1 (max) the most expensive at $13.04
0
0
3
192
Raphi-2Code retweeted
GPT-6.1 Sol beats Grok 4.7 on FrontierCode 2.0 for 1/14 the price
Grok 4.8 wen? I hear it may even be GPT 6 Sol
Grok 4.8 wen? I hear it may even be GPT 6 Sol
0
1
10
1,051
@Noah_Twenty and Sol should’ve been called Terra due to the similar pricing
0
0
1
26