Key takeaways of zdnet
- Anthropic launched Cloud Opus 4.1.
- The model is higher than the performance of the anterior on complex functions.
- This paid is available for Cloud users, cloud codes, API, Amazon Bedrock and Vertex AI of Google Cloud.
In May, Anthropic released Cloud Oppus 4, which the company dubbed its most powerful model and the best coding model in the world. Only three months later, the anthropic is moving ahead by launching the much awaited cloud Opus 4.1, which now takes the crown of its predecessor as the most advanced model of anthropic.
The model’s opas family is the most advanced, intelligent AI model of the company that is leading to dealing with complex problems. Consequently, Cloud Opus 4.1, released on Tuesday, excellence on those tasks and even according to anthropic, can make its predecessor on agentic functions, real-world coding and logic.
The model also comes as the industry is expecting Openai’s launch of GPT-5 soon.
Also: Openai can now launch any minute GPT -5 – what is expected
How does Cloud Opus 4.1 perform?
One of the most impressive use cases of Cloud Oppus 4 was that the self-bench was verified, a human-filtated most of the self-bench, its performance on a benchmark, which evaluates the capabilities of LLMS to solve software engineering works of the real-world software engineering obtained from GITHUB. The performance of Cloud Ops 4 on Swe-Bench verified supported the claim that it was the “best coding model in the world”. As seen in the above post, Ops 4.1 performed even more.
Cloud Opus 4.1 also excluded its predecessor model in the benchmark board including MMLU, which tests for multilingual abilities; AIME 2025, which tests for rigidity on high school match competition questions; GPQA, which tests for performing on the signal of graduate level logic; More. When pinned against contestants’ logic models, including Open O3 and Gemini 2.5 Pro, it improves them in various benchmarks, including SWE-Bench verified.
With release, Anthropic also posted it System cardWhich portrays all the safety assessment and evaluation made on the model, as well as its weaknesses, risks and limitations. A quick observation of the 22 -Rashtriya document suggests that the model was deployed with the AI security level 3 (ASL -3) standard under the athropic scaling policy (RSP), and is still prone to most similar weaknesses.
How can you reach Cloud Opus 4.1?
If you want to try a model for yourself, it is now available to all through paid cloud schemes, including $ 20 per month for $ 20 per month for $ 100 per month. It is available in Cloud Code, API, Amazon Bedrock and Vertex AI of Google Cloud.
Get top stories of morning with us in your inbox every day Tech Today Newsletter.
