Anthropic发布Claude Opus 5并接入AWS、微软、Databricks和GitHub，亚洲云AI与集成商映射缺少可执行边际

# 美股盘后催化 9/11: Anthropic发布Claude Opus 5并登陆AWS、Microsoft Foundry、Databricks和GitHub，企业AI竞争转向成本/性能

- 美股会话: 2026-07-24
- 方向: MIXED；类型: 产品/技术；规划重大性: 7/10
- 映射模式: competitive
- 美股驱动（规划器初判，需独立核验）: 事实：Anthropic发布Claude Opus 5，称在编码、知识工作和科研等任务上提升，接近Fable 5且成本约为一半；AWS Bedrock、Microsoft Foundry、Databricks、GitHub Copilot等同步接入。推断：这利好企业AI采用、云AI平台和开发工具生态，但加剧模型价格竞争，也可能压缩高价模型和算力单token收入假设。

## 本催化相关美股真实收盘行情

- 英伟达 (NVDA.US): 206.84 -0.92%
- AMD (AMD.US): 521.95 -3.29%
- 微软 (MSFT.US): 381.7 +0.03%
- 谷歌 (GOOGL.US): 319.74 +0.65%
- 亚马逊 (AMZN.US): 232.11 -0.66%

## 亚太映射假设（仅候选，必须查证）

- 阿里巴巴 (9988.HK) [HK/竞争替代/mixed]: 企业AI模型性价比竞争升级，阿里云需要证明模型能力、价格和企业集成不被海外平台拉开。；核验问题: 核实阿里云同类模型定价、企业客户留存、推理毛利和海外可用性。
- 腾讯控股 (700.HK) [HK/竞争替代/mixed]: 海外强模型降价推动企业AI采用，但也提高腾讯混元和企业AI产品的价格/能力压力。；核验问题: 核实腾讯企业AI产品定价、客户留存和模型供应路径。
- NTT Data (9613.JP) [JP/客户/集成商/positive]: 系统集成商可把Bedrock/Azure上的前沿模型纳入企业AI项目，提高项目交付能力。；核验问题: 核实NTT Data与AWS/Azure AI服务的合作和收入敞口。

## 必答问题

- Opus 5实际API调用量是迁移还是新增？
- 模型降价是扩大推理量还是压低云AI毛利？
- 亚洲互联网和SaaS公司能否合规接入这些模型？

## 原始证据全文（14 条）

### [E1109] Anthropic | 2026-07-24 17:00 UTC | 盘中
Introducing Claude Opus 5
Claude Opus 5 is available today. It’s a thoughtful and proactive model that comes close to the frontier intelligence of Claude Fable 5 at half the price.
On coding and knowledge work evaluations like Frontier-Benc
URL: https://www.anthropic.com/news/claude-opus-5
Introducing Claude Opus 5
Claude Opus 5 is available today. It’s a thoughtful and proactive model that comes close to the frontier intelligence of Claude Fable 5 at half the price.
On coding and knowledge work evaluations like Frontier-Bench and GDPval-AA, Opus 5 is the new state-of-the-art, though it remains behind Mythos 5 on cybersecurity tasks.
Opus 5 is designed to be used every day: it works more efficiently than other models. It’s the new default model on Claude Max, and the strongest model on Claude Pro.
Performance and cost-effectiveness
Claude Opus 5 provides greatly improved performance for the same cost as its predecessor, Opus 4.8. The charts in this section show how performance changes according to the model’s effort setting, which customers can use to optimize for intelligence or conserve tokens for faster and cheaper results.
Opus 5 excels on valuable software engineering tasks. For example, on Frontier-Bench v0.1, Opus 5 surpasses all other models, and more than doubles Opus 4.8’s performance at a lower cost per task. On CursorBench 3.2, at max effort, the model performs within 0.5% of Fable 5’s peak score, but at half the cost per task; it also achieves greater performance at a given cost than all other models on high, xhigh, and max effort.
We see similar results on knowledge work and problem-solving tasks. For example:
- On ARC-AGI 3, an evaluation where the model has to solve novel problems, Opus 5’s score is three times as high as the next-best model.
- On Zapier AutomationBench, which measures whether models can complete business tasks from start to finish, Opus 5’s pass rate is around 1.5× the next-best model for the same cost per task. Even at its lowest effort setting, Opus 5 passes more tasks than any other model.
- On OSWorld 2.0, a computer use benchmark, Opus 5 outperforms every other model at any given cost, surpassing Fable 5’s best result at just over a third of the cost.
It’s also our best and most cost-efficient model on several related evaluations:
Opus 5 is a meaningful improvement over Opus 4.8 for scientific research. It shows better performance than Opus 4.8 on every one of our life sciences evaluations, which cover topics including structural biology, organic chemistry, and bioinformatics. Its improvements are most notable on organic chemistry tasks, like inferring molecular structures from spectroscopy data (it scores 10.2 percentage points higher than Opus 4.8 on our internal benchmark), and on protein-related tasks like predicting how variations in a protein’s sequence affect how it functions (here, it scores 7.7 percentage points higher).
Finally, Opus 5 is capable of producing much stronger visual outputs:
Working with Claude Opus 5
Claude Opus 5 is much stronger at verifying its work and iterating carefully until it succeeds. In evaluations and early-access testing, we and our users found many examples of Opus 5’s agency and thoroughness:
- On one Frontier-Bench task, Opus 5 was given a drawing of a machine part and asked to write code to rebuild it as a 3D FreeCAD model. However, in this task, the model was intentionally given no way to directly view the drawing. Opus 5 responded by writing its own computer vision pipeline to pull the geometry from the raw pixels, then reconstructed the full machine part. It succeeded in doing so repeatedly; no competing model with the same setup could solve it after five attempts.
- Given a real bug in a popular open-source package manager, Opus 5 found the root cause and fixed an edge case that the community’s patch had missed. A competing model fixed only the surface symptom (not the underlying cause), then reported the bug resolved.
- An engineer at a trading firm used Opus 5 to build a market data feed for a new exchange in a single session. Previous models could not complete this task at all, even given extensive plans from the engineer. Finding no live feed to validate against, Opus 5 even built its own test harness to check that its code parsed the exchange’s data correctly.
Below are further reports from our early-access customers on their experience of working with Opus 5:
On FrontierCode 1.1, Claude Opus 5 approaches Fable-level performance at half the cost. Within Devin, it also shows particular strength on difficult debugging and root-cause analysis tasks.
Claude Opus 5 delivers near Fable 5 intelligence at Opus speed and cost. On CursorBench it’s just under Fable 5 and has many of the same behaviors. We are excited to see how developers use it in Cursor.
Claude Opus 5 topped Zapier’s AutomationBench leaderboard without spending more tokens than prior Claude models. It took a raw account-health workbook and ran a full churn-prevention sequence end to end: flagging at-risk accounts, alerting the right owner, and summarizing for retention ops. Previous models didn’t pass; Opus 5 hit 100%.
On our genomics analysis work, Claude Opus 5 behaves more like a careful scientist than any model we’ve run. It reaches for the right statistical tests to rule out confounders, cross-checks its own results by independent methods, and stays on track through long multi-step analyses.
Claude Opus 5 came out ahead of every model in its family on our internal evals. It isn’t just better on our hardest agentic coding tasks, up 22% over Opus 4.7, it’s steadier, with far less variance run to run. For the millions of builders on Lovable, that consistency is the whole game. Reliable results, build after build.
Claude Opus 5 is the biggest leap in the Opus family since 4.5. On the same full-stack app builds, the front end shows it first: the best animations, games, and 3D work we have seen from an Opus model.
We’re loving Claude Opus 5. For the kind of open-ended analytical work our agent handles, it’s a strict upgrade over Opus 4.8, and the gains are biggest exactly where it matters: the harder, vaguer tasks. Responses are clearer and more concise, and we see improved efficiency at higher effort levels too.
Claude Opus 5 is a striking improvement over Opus 4.8 for the financial research workflows our analysts run every day. It stands out on numerical reasoning, table work, and sharper critical thinking where precision matters.
Claude Opus 5 delivers the industry intelligence and accuracy that is essential for the analysis of specialized enterprise content. Box found that Opus 5 outperforms Opus 4.8 by 8% and delivers notable performance gains in the data analysis (11% improvement) and due diligence (17% improvement) workflows that technology, healthcare, and public sector organizations rely on daily.
Claude Opus 5 is a clear generational step up from Opus 4.8. Over one weekend I gave it a chief-of-staff role over my dev environments: it built its own monitor, drove each box, and pulled me in only for the judgment calls.
Claude Opus 5 made large scale changes across our Fundamental Research Assistant codebase, adapting to feedback throughout an agentic workflow and explaining its reasoning more clearly than any model we’ve used. It handled work we would normally have broken into much smaller pieces.
On some of our hardest financial-modeling tasks, Claude Opus 5 is a clear step up from Opus 4.8 in both accuracy and efficiency. Its performance floor is materially higher, especially on deep finance domain logic. Across effort levels it averaged 9 percentage points higher accuracy with a third fewer turns and tool calls and 60% less time.
Claude Opus 5 checks its own work the way a real frontend developer would. On our benchmark it opened its pages in a browser at desktop and phone widths, caught a product hidden below the mobile fold and an off-screen checkout button, and fixed both before handing the work back.
Claude Opus 5 is a clear step up in performance on legal agent work compared to prior Opus models, and we saw the biggest gains in practice areas like corporate governance and arbitration. We were also impressed with Opus 5’s ability to maintain quality at lower reasoning levels, achieving similar performance while generating 26% fewer tokens on average compared to Opus 4.8 at max reasoning.
Claude Opus 5’s biggest gains for us are on longer-horizon work: building a full deck, then revising it. Artifact quality is what decides which model we ship, and this is the clearest step up we’ve seen — better visual understanding, cleaner formatting, fewer slide issues.
Claude Opus 5’s judgment is what stands out. Handing off a PR, it doesn’t rush to publish: it verifies the branches, checks the template, and thinks through test implications so the handoff is clean. The older models tended to jump ahead and get caught on our checks.
During a rearchitecting session, Claude Opus 5 pushed back on a design I proposed, and it didn’t fold when I insisted. Instead, it explained exactly what was valuable in my idea, narrowed its objection to a single design question, and proposed a compromise that kept the good part while fixing the flaw. That’s the kind of judgment that lets us trust it with less oversight.
On first-turn redlines, Claude Opus 5 scored the highest of any model we tested, nearly double Opus 4.8. Commenting is better too: on NDAs it gets to the redline in less time and with fewer passes, with accuracy maintained or better.
Claude Opus 5 writes clean, tight diffs with no dead code, and it’s the stronger hazard spotter on subtle, codebase-specific issues. We’re adopting it for production workloads.
We will definitely migrate a number of use cases in Cosmos, our unified agent platform. We’re looking forward to increasingly using Claude Opus 5 for code review, and I am confident in saying we would rather people be using Opus 5 than Opus 4.8.
What stands out about Claude Opus 5 is judgment. It thinks harder before it writes a single line, catches its own logical faults during planning rather than after the fact, and reasons about why an answer is right, not just whether it works. It’s the clearest jump in problem-solving we’ve seen from one Claude model to the next, and we’re looking forward to seeing it adopted in JetBrains IDEs.
Claude Opus 5 is the strongest Opus model we’ve tested on our trading benchmark, and it gets there using roughly a seventh of the reasoning tokens and under half the latency of Opus 4.8. Better answers at a fraction of the compute.
Alignment and safety
Alignment. During pre-deployment testing, our automated behavioral audit found Opus 5 to be our most aligned model to date (as shown in the graph below). It adheres to Claude’s Constitution better than Opus 4.8, Sonnet 5, or Fable 5; exhibits the lowest rates of deceptive behavior; and is the least susceptible to being tricked into misuse. It’s also our safest model yet in terms of avoiding reckless actions that could have hard-to-reverse side effects.
Safety. Opus 5 does not advance the frontier in risky, dual-use capabilities. In rigorous evaluations conducted alongside private-sector and government partners, we found it remains behind Mythos 5 in both biology research and offensive cybersecurity. More information about these evaluations can be found in our System Card.
As with its predecessor, Opus 4.8, we’ve intentionally avoided training Opus 5 on cyber tasks. The model has nevertheless improved substantially on these tasks as a result of becoming more generally capable, and it comes close to Mythos 5 at finding cybersecurity vulnerabilities. However, it remains substantially behind Mythos 5 on the exploitation of those vulnerabilities—that is, in turning vulnerabilities into material cyber threats.
This is illustrated by Opus 5’s performance on OSS-Fuzz, an evaluation we’ve developed to assess how well models can find and then exploit vulnerabilities without extensive human guidance. Although Mythos 5 and Opus 5 identify vulnerabilities with similar success, Opus 5’s score on the development of exploits is far behind that of Mythos 5.
Safeguards for Opus 5
Claude Opus 5’s safeguards are designed to allow beneficial uses of the model in both cybersecurity and biology. They are similar to those we applied to Opus 4.8, with the exception of some stronger guardrails on a narrow range of cyber tasks.
Cybersecurity. Opus 5’s cyber classifiers are proportionally less restrictive than those on Fable 5. They allow Opus 5 to find vulnerabilities in source code, but block “binary-based” vulnerability scanning (a method more likely to be associated with malicious actors), penetration testing, and exploit generation.
Based on our testing, we expect the classifiers to intervene around 85% less often than they do for Fable 5. In Claude.ai, Claude Code, and Claude Cowork, any flagged requests will fall back to Opus 4.8 by default. Fallbacks to Opus 4.8 can also be enabled on the API.
Our Cyber Verification Program (CVP) facilitates cybersecurity work that would otherwise be impeded by the model’s safeguards. Enterprises and researchers who are already part of the CVP have immediate access to a version of Opus 5 with fewer security restrictions.
Biology. Since Opus 5 has a similar suite of safeguards to Opus 4.8, it is now our most capable generally available model for scientific research. Nevertheless, the model still shows important limitations on long-running, autonomous research tasks, which is where we expect AI models to pose the most substantial biology-related risks. (Mythos 5 remains the stronger model for this type of biological work.) As part of this launch, biology-related requests that are blocked on Fable 5 will now route to Opus 5 rather than Opus 4.8.
Getting started
Claude Opus 5 is available today on all platforms, priced at $5 per million input tokens and $25 per million output tokens (the same as Opus 4.8). Developers can get started with claude-opus-5 on the Claude API.
It’s also offered in Fast mode, where it runs around 2.5 times the default speed. As with Opus 4.8, Fast mode is available at twice Opus 5’s base price on the Claude Platform and through usage credits in Claude Code.
Alongside Opus 5, we’re releasing two updates in beta:
- Mid-conversation tool changes on the Claude Platform. Within a conversation, developers can now change which tools Claude can use without invalidating the prompt cache.
- Automatic fallbacks on the API. Users can now choose to have requests that are flagged by our safety classifiers on Opus 5 (or Fable 5) automatically route to another model. With automatic fallbacks on, API requests always route to the best available model by default rather than being blocked.
Consistent with prior Opus models, Opus 5 does not have data retention requirements for general access.
For more guidance on how to get the best out of Opus 5, see our prompting guide.
Footnotes
Frontier-Bench v0.1, Effort plot: These results are from an internal run of Frontier-Bench v0.1, on the mini-SWE-agent harness and a GKE backend, mean reward over 5 attempts per task. Opus 4.8 served as fallback on safety-classifier refusals for Opus 5 and Fable 5.
Related content
A research agenda for the Economic Futures Research Fund
We’re sharing the research agenda for the Anthropic Economic Futures Research Fund.
Read more

### [E1113] Bloomberg | 2026-07-24 17:00 UTC | 盘中
Anthropic Unveils More Cost-Efficient Model for Everyday Tasks
URL: https://www.bloomberg.com/news/articles/2026-07-24/anthropic-unveils-more-cost-efficient-model-for-everyday-tasks
Anthropic PBC is rolling out a new artificial intelligence model that is meant to handle workplace tasks adeptly and affordably, a key selling point as some customers grow more cost conscious and competition increases in China.

The company on Friday unveiled Claude Opus 5, a model it says approaches the capabilities of its most advanced widely available offering, Fable 5, in many categories but at half the price. Anthropic said it expects Opus 5 to serve as the default option for many day-to-day office needs.

A number of AI developers, including OpenAI, Meta Platforms Inc. and SpaceXAI, are competing to offer more cost-efficient AI models as businesses increasingly scrutinize their AI spending. Some customers have recently imposed tighter limits on employee AI usage after being hit by sticker shock.

“We’re very much responding to this feedback from enterprises about, ‘How do I generate more value?’” said Dianne Penn, head of research product management at Anthropic.

The emphasis on affordability may grow as China shows signs of gaining ground on the US in the AI race. Last week, Chinese AI lab Moonshot released Kimi K3, a more advanced open-weight model that it said outperforms all rivals except for Fable 5 and OpenAI’s GPT-5.6 on overall capability. Other Chinese firms, including DeepSeek, have pushed to undercut US rivals on pricing.

Anthropic said Opus 5 is a general-purpose model with improved abilities to check its own work, field scientific research and streamline coding, a lucrative market for AI. The model is also nearly as capable as Anthropic’s more advanced Mythos 5 offering in finding cybersecurity vulnerabilities, but “remains substantially behind” in exploiting them, Anthropic said.

“As with its predecessor, Opus 4.8, we’ve intentionally avoided training Opus 5 on cyber tasks,” the company said in a blog post. “The model has nevertheless improved substantially on these tasks as a result of becoming more generally capable.”

Anthropic previously delayed the release of Fable 5 and Mythos 5 under pressure from the US government. It later redeployed Fable 5 with a new set of classifiers to target and block more cybersecurity tasks.

The company said Opus 5’s classifiers are “proportionally less restrictive” than with Fable 5, meaning those safeguards will intervene significantly less often and block a more limited range of tasks. Anthropic said it conducted “rigorous evaluations” of the model “alongside private sector and government partners.”

Rival OpenAI also delayed the introduction of its most recent model, GPT-5.6, while the government reviewed it. Earlier this week, OpenAI said its most advanced models inadvertently breached AI startup Hugging Face’s internal systems, an incident that the ChatGPT maker described as “unprecedented.” The hack was carried out in a matter of hours.

Penn said OpenAI’s breach is a reminder of AI’s rapidly improving capabilities and the need for companies to be “continuously updating our evaluations and our security systems.”

### [E1115] CNBC Top News | 2026-07-24 17:00 UTC | 盘中
Anthropic's new AI model rivals Fable 5 and is cheaper as businesses fret about costs
URL: https://www.cnbc.com/2026/07/24/anthropic-claude-opus-5-ai-fable-5-cost.html
Anthropic on Friday announced its latest artificial intelligence model, Claude Opus 5, which it said is its best-performing and most cost-effective offering across a number of industry benchmarks.
Opus 5 outperforms the last model that Anthropic released, Claude Fable 5, on coding and knowledge work evaluations, and the company said it's "designed to be used every day," according to a blog post. Opus 5 is also half the price of Fable 5, and will cost users $5 per million input tokens and $25 per million output tokens.
Anthropic said Opus 5 is not the state-of-the-art for "risky, dual-use capabilities," like cybersecurity.
The model launch comes as Anthropic and its chief rival, OpenAI, have been working to appease a more cost-conscious customer base, where enterprises are less inclined to experiment with pricey models without a clear picture of the return on their investment. Anthropic is also facing pressure from competitors that tout cheaper offerings, including Microsoft, Amazon, Google, and a number of Chinese startups.
"Enterprises, in our feedback and with our customer base, are looking for value," Dianne Penn, Anthropic's head of product management for research, told CNBC in an interview. "If it's a cheaper model or a cheaper offering, but it's not accomplishing a similar level of quality, it's actually not useful."
Anthropic has been the subject of intense scrutiny since it released a model called Claude Mythos Preview in April, which it rolled out to a select group of users. The company said the model was extremely capable in cybersecurity, and that it excelled at discovering and exploiting vulnerabilities within software.
In June, Anthropic released an updated version of the model, Mythos 5, and it also launched another version of the model, Fable 5, that it said was safe for general use. The company said Fable 5's capabilities exceeded those of all other models it had ever released publicly, according to a blog post.
Anthropic had to briefly disable access to Mythos 5 and Fable 5 days after they were released in order to comply with an export control directive from the government that cited "national security authorities." The export controls were lifted after roughly two weeks of intense negotiations, and Anthropic was able to restore access to both models.
Penn said Anthropic is continuing to collaborate with government agencies as part of its pre-deployment and testing, including with Opus 5. The company said Opus 5 remains behind Mythos 5 in offensive cybersecurity and biology research.
WATCH: Anthropic partnership is driving uptick in Google could revenue, says DA Davidson’s Gil Luria

### [E1142] Reuters | 2026-07-24 17:02 UTC | 盘中
Anthropic rolls out Opus 5 AI model in efficiency upgrade
URL: https://www.reuters.com/technology/anthropic-rolls-out-opus-5-ai-model-efficiency-upgrade-2026-07-24/
Anthropic on Friday launched Opus 5, its latest AI model that the startup says nears the capabilities of ​its more powerful cousin Fable 5 at half the price.

The ​San Francisco-based lab said the new Claude AI model ⁠was well suited for daily office and computer programming tasks.

In an ​interview with Reuters, Anthropic product leader Dianne Penn said the release, ​more efficient than May's Opus 4.8, reflected a rapid pace of development.

"We're building and continue to consistently deliver frontier intelligence and bring that as accessibly as ​possible with every model generation," said Penn.

In testing, Opus 5 ​was less capable of exploiting cyber vulnerabilities than Anthropic's top-shelf AI, so its ‌related ⁠safeguards are less restrictive than Fable 5's. Opus 5 was also less susceptible to being tricked into misuse than Anthropic's other current models, the startup said.

Released in June, Anthropic's Fable 5 was temporarily unavailable ​following U.S. concerns that ​its capabilities ⁠could be diverted to foreign military intelligence.

Users should pick Opus 5 for value and Fable 5 for "days-long, very ​autonomous projects," Penn said.

Asked about the Kimi K3 "open" model ​from China-based ⁠Moonshot, which the U.S. accused of freeloading off Anthropic, Penn said it "remains to be seen" how open-weight models generally perform on complicated real-world ⁠projects that ​users ask Claude to tackle.

Open-weight models ​allow users to download, run and customise the virtual brains of an AI, unlike proprietary ​models.

### [E1175] GitHub | 2026-07-24 17:12 UTC | 盘中
📣 @AnthropicAI's Claude Opus 5 is now available and rolling out in GitHub Copilot.

Early testing shows
➡️ It has strong performance on agentic coding workflows
➡️ It's effective at making targeted changes, validating its work, and reducin
URL: https://x.com/github/status/2080702791887315229
📣 @AnthropicAI's Claude Opus 5 is now available and rolling out in GitHub Copilot.

Early testing shows
➡️ It has strong performance on agentic coding workflows
➡️ It's effective at making targeted changes, validating its work, and reducing unnecessary execution overhead on complex tasks

Try it out in the GitHub Copilot app or @code 👇

### [E1238] Google Cloud Tech | 2026-07-24 17:34 UTC | 盘中
Claude Opus 5 is live on Agent Platform!

Build and scale secure AI agents with model choice and flexibility. Try it today → https://goo.gle/4gVWDIy
URL: https://x.com/GoogleCloudTech/status/2080708199594823736
Claude Opus 5 is live on Agent Platform!

Build and scale secure AI agents with model choice and flexibility. Try it today → https://goo.gle/4gVWDIy

### [E1254] Databricks | 2026-07-24 17:39 UTC | 盘中
.@AnthropicAI's Claude Opus 5 is now available where your data already lives - governed, secure, and ready for custom AI applications and agents built with the data in your Lakehouse.

On Databricks' internal coding bench, Opus 5 achieved S
URL: https://x.com/databricks/status/2080709560726163764
.@AnthropicAI's Claude Opus 5 is now available where your data already lives - governed, secure, and ready for custom AI applications and agents built with the data in your Lakehouse.

On Databricks' internal coding bench, Opus 5 achieved SOTA performance. It delivers major gains in agentic coding, professional knowledge work, and long-horizon reasoning, with stronger performance per token across effort levels. Teams can build agents that handle complex workflows with fewer iterations and meaningfully lower cost, all governed by Unity AI Gateway

Put Claude Opus 5 to work on your most important data, securely and at production scale.

### [E1268] Microsoft Azure | 2026-07-24 17:43 UTC | 盘中
Meet Claude Opus 5, Anthropic's most advanced Opus model now in Microsoft Foundry. Combined with Microsoft Foundry, organizations can evaluate, build, deploy, and scale enterprise AI with native security controls, governance, and data contr
URL: https://x.com/Azure/status/2080710486937534673
Meet Claude Opus 5, Anthropic's most advanced Opus model now in Microsoft Foundry. Combined with Microsoft Foundry, organizations can evaluate, build, deploy, and scale enterprise AI with native security controls, governance, and data controls.

Learn more:

### [E1293] Seeking Alpha | 2026-07-24 17:50 UTC | 盘中
Anthropic launches Opus 5, closer and cheaper to frontier AI model Fable 5
URL: https://seekingalpha.com/news/4618312-anthropic-launches-opus-5-closer-and-cheaper-to-frontier-ai-model-fable-5?utm_source=feed_news_all&utm_medium=referral&feed_item_type=news
On Friday, Anthropic (ANTHRO) launched its new AI model Claude Opus 5, which comes close to its frontier AI model Claude Fable 5 at half the price.
The AI tech giant noted that Opus 5 is priced the same as Opus 4.8, with materially higher performance across coding, knowledge work, and scientific research benchmarks.
Opus 5 is priced at $5 per million input tokens and $25 per million output tokens, the same as Opus 4.8, which was unveiled in May. Developers can get started with Claude Opus 5 on the Claude API, according to the company.
The new model is also offered in Fast mode, where it runs around 2.5 times the default speed. As with Opus 4.8, Fast mode is available at twice Opus 5's base price on the Claude Platform and through usage credits in Claude Code.
Anthropic noted that Claude Opus 5's safeguards are designed to allow beneficial uses of the model in both cybersecurity and biology. They are similar to those applied to Opus 4.8, except for some stronger guardrails on a narrow range of cyber tasks.
Anthropic is in a tight race with rival OpenAI (OPENAI) to develop advanced AI models. However, managing these models comes with its own set of challenges. U.S. lawmakers have proposed an "AI Kill Switch Act," which would allow federal authorities to halt AI models. This comes just days after OpenAI said its AI agent escaped containment during a security test.
Last month, Anthropic was forced to temporarily restrict access to its most advanced AI models Fable 5 and Mythos 5 after the Trump administration issued an export control directive. OpenAI had also staggered the release of GPT-5.6 following a government request.
Opus 5’s cyber classifiers are proportionally less restrictive than those on Fable 5. They allow Opus 5 to find vulnerabilities in source code, but block 'binary-based' vulnerability scanning — a method more likely to be associated with malicious actors — penetration testing, and exploit generation, according to the company.
The company added that since the new model has a similar suite of safeguards to Opus 4.8, it is now its most capable generally available model for scientific research. However, the model still has limitations on long-running, autonomous research tasks.
Mythos 5 remains the stronger model for some types of biological work, the company noted.
In addition, Opus 5 is much stronger at verifying its work and iterating carefully until it succeeds.
The model also excels on software engineering tasks. On Frontier-Bench v0.1, Opus 5 surpasses all other models and more than doubles Opus 4.8’s performance at a lower cost per task. On CursorBench 3.2, at max effort, the model performs within 0.5% of Fable 5’s peak score, but at half the cost per task, Anthropic added.

### [E1309] AWS Machine Learning Blog | 2026-07-24 17:59 UTC | 盘中
Today, we announce the availability of Claude Opus 5 on Amazon Bedrock and Claude Platform on AWS. Claude Opus 5 is Anthropic’s most advanced Opus model and the first in the fifth generation. It is a meaningful step forward, providing impro
URL: https://aws.amazon.com/blogs/machine-learning/introducing-claude-opus-5-on-aws-anthropics-most-capable-opus-model/
Today, we announce the availability of Claude Opus 5 on Amazon Bedrock and Claude Platform on AWS. Claude Opus 5 is Anthropic’s most advanced Opus model and the first in the fifth generation. It is a meaningful step forward, providing improvements across the workflows that teams run in production such as agentic coding, knowledge work, visual understanding, and long-running tasks. According to Anthropic, Claude Opus 5 matches Claude Fable 5’s top-tier intelligence in many domains at Opus-tier pricing.

On Amazon Bedrock, Claude Opus 5 benefits from zero data retention (ZDR) by default, delivering top-tier intelligence without compromising enterprise data governance. Claude Opus 5 is powered by Bedrock’s next-generation inference engine, letting you build within your existing AWS environment, maintain enterprise security and regional data residency, and scale inference with zero operator access.

Claude Opus 5 is also available through Claude Platform on AWS, which supports zero data retention on request and gives you access to Anthropic’s native platform experience and capabilities through the AWS Console. Build, test, and deploy with the same APIs, features, and console experience you get working with Anthropic directly, unified with AWS billing and authentication.

This post covers Opus 5’s improvements and practical guidance for AI engineers integrating the model into agentic systems and production inference workloads on Amazon Bedrock. See the documentation for Claude Platform on AWS.

What makes Claude Opus 5 different

According to Anthropic, Claude Opus 5 delivers a step-change in coding. It understands and navigates codebases like an experienced engineer and writes production-quality code while adapting its strategy as it works. It powers dependable long-running agents that work for hours and even overnight, finding paths around obstacles, recovering from errors, and reaching their objectives. For professional work, it brings deeper reasoning to long documents and higher accuracy to complex analysis, with the largest gains on document-heavy enterprise tasks. Together, this adds up to frontier capability at Opus-tier economics.

Use cases

Claude Opus 5 is a strong fit for industries where precision, reliability, and deep reasoning matter most. For financial services teams, it powers financial workflows with deeper reasoning and end-to-end understanding of context. It handles compliance-sensitive work with more clarity and focus. For agent and workflow automation, it pushes back on flawed instructions and breaks complex jobs into sub-agents that need less oversight. In productivity work, it handles report building and auditing, document drafting, and structured analysis with high consistency. It runs multi-day projects and produces professional-grade outputs.

Claude Opus 5 also improves on Claude Opus 4.8’s cyber capabilities across the board, from coding to cyber security. In higher-risk areas, Opus 5 may fall back to Opus 4.8. Users are notified when this happens, and fallbacks can be configured when using the APIs.

Getting started with Claude Opus 5 on Amazon Bedrock

You can get started with Claude Opus 5 in the Amazon Bedrock console .

In the Amazon Bedrock console, under Test, choose Playground.

For the model, choose Claude Opus 5. Now, you can test your complex coding prompt with the model.

Amazon Bedrock console Playground with Claude Opus 5 selected

You can also access the model programmatically using the Anthropic Messages API to call bedrock-runtime through the Anthropic SDK or bedrock-mantle endpoints, or keep using the Invoke and Converse API on bedrock-runtime through the AWS Command Line Interface (AWS CLI) and AWS SDK .

Prerequisites

Active AWS account with Amazon Bedrock access.

AWS CLI installed and configured.

Python 3.8+.

Boto3 installed: pip install boto3

Anthropic SDK installed: pip install anthropic[bedrock]

IAM permissions: bedrock:InvokeModel , bedrock:InvokeModelWithResponseStream , and bedrock:CreateInference

Here’s a quick example using the AWS SDK for Python (Boto3):

import boto3
import json

# Create a Bedrock Runtime client
bedrock_runtime = boto3.client(
service_name="bedrock-runtime",
region_name="us-east-1"
)

# Invoke Claude Opus 5
response = bedrock_runtime.invoke_model(
modelId="global.anthropic.claude-opus-5",
contentType="application/json",
accept="application/json",
body=json.dumps({
"anthropic_version": "bedrock-2023-05-31",
"max_tokens": 4096,
"messages": [
{
"role": "user",
"content": "Design a distributed architecture on AWS in Python that should support 100k requests per second across multiple geographic regions."
}
]
})
)

result = json.loads(response["body"].read())
print(result["content"][1]["text"])

You can also use Claude Opus 5 with the Amazon Bedrock Converse API for a unified multi-model experience:

import boto3

bedrock_runtime = boto3.client("bedrock-runtime", region_name="us-east-1")

response = bedrock_runtime.converse(
modelId="global.anthropic.claude-opus-5",
messages=[
{
"role": "user",
"content": [
{
"text": "Design a distributed architecture on AWS in Python that should support 100k requests per second across multiple geographic regions."
}
]
}
],
inferenceConfig={
"maxTokens": 4096
}
)

if 'output' in response:
blocks = response['output']['message']['content']
print('\n'.join(b.get('text', '') for b in blocks if 'text' in b))

You can also use Claude Opus 5 with the Anthropic Messages API using the anthropic[bedrock] SDK package for a streamlined experience:

from anthropic import AnthropicBedrockMantle

# Initialize the Bedrock Mantle client (uses SigV4 auth automatically)
mantle_client = AnthropicBedrockMantle(aws_region="us-east-1")

# Create a message using the Messages API
message = mantle_client.messages.create(
model="anthropic.claude-opus-5",
max_tokens=4096,
messages=[
{"role": "user", "content": "Design a distributed architecture on AWS in Python that should support 100k requests per second across multiple geographic regions"}
]
)

print(message.content[1].text)

Claude Opus 5 supports adding and removing tools mid-conversation through tool_addition and tool_removal content blocks on role: "system" messages, instead of re-sending the full top-level tools array. See Bedrock documentation for more details.

Availability

Claude Opus 5 is available today on Amazon Bedrock in Regions including US East (N. Virginia), Asia Pacific (Melbourne), Europe (Ireland), and Europe (Stockholm). See Bedrock documentation for the full list of supported Regions. Claude Opus 5 is also available on the Claude Platform on AWS in North America, South America, Europe, and Asia Pacific.

Give Claude Opus 5 a try in the Amazon Bedrock console , in Claude Platform on AWS , or explore the Getting Started notebooks on GitHub. You can also unlock the full potential of Opus 5 by using Advanced Prompt Optimization on Amazon Bedrock. It takes your current prompts, benchmarks them against your evaluation criteria, and outputs production-ready rewrites.

About the authors

### [E1337] 新浪财经7x24 | 2026-07-24 18:03 UTC | 盘中
亚马逊云科技（AWS）：宣布 Claude Opus 5 现已登陆亚马逊 Bedrock 平台，同时 AWS 上线 Claude 专属平台。
URL: https://wap.cj.sina.cn/pc/7x24/5007451
亚马逊云科技（AWS）：宣布 Claude Opus 5 现已登陆亚马逊 Bedrock 平台，同时 AWS 上线 Claude 专属平台。

### [E1340] 财联社电报 | 2026-07-24 18:05 UTC | 盘中
财联社7月25日电，AWS宣布在Amazon Bedrock和AWS上接入Claude Opus 5。
URL: https://www.cls.cn/detail/2436836
财联社7月25日电，AWS宣布在Amazon Bedrock和AWS上接入Claude Opus 5。

### [E1400] AWS Newsroom | 2026-07-24 18:25 UTC | 盘中
.@AnthropicAI's most advanced Opus model, Claude Opus 5, is now available on AWS. Improvements across agentic coding, autonomous tasks, and complex knowledge work, with zero data retention compatibility for enterprise workloads. https://go.
URL: https://x.com/AWSNewsroom/status/2080721167761969204
.@AnthropicAI's most advanced Opus model, Claude Opus 5, is now available on AWS. Improvements across agentic coding, autonomous tasks, and complex knowledge work, with zero data retention compatibility for enterprise workloads. https://go.aws/4vJoDTD

### [E1149] 研报平权 | 2026-07-24 17:03 UTC | 盘中
英国《金融时报》：“假设每天工作八小时，AI采用者的平均使用时长已从2024年末的大约20分钟，增至2026年年中的约30分钟。大约三分之一的AI使用是为了节省时间。这意味着，每个工作日节省的时间已从2024年的约七分钟增至2026年的约10分钟。放大到一个国家的劳动力规模，这是一项颇具价值的效率提升。目前尚不清楚的是，劳动者会把每天额外获得的2%工作时间用于提高整体生产率、核查并修正AI生成的内容，还是用来偷懒。”
这些数据还牵涉到一个问题：人们感知到的AI时间节省，是否是
URL: https://wx.zsxq.com/dweb2/index/topic_detail/82255445511414822
英国《金融时报》：“假设每天工作八小时，AI采用者的平均使用时长已从2024年末的大约20分钟，增至2026年年中的约30分钟。大约三分之一的AI使用是为了节省时间。这意味着，每个工作日节省的时间已从2024年的约七分钟增至2026年的约10分钟。放大到一个国家的劳动力规模，这是一项颇具价值的效率提升。目前尚不清楚的是，劳动者会把每天额外获得的2%工作时间用于提高整体生产率、核查并修正AI生成的内容，还是用来偷懒。”
这些数据还牵涉到一个问题：人们感知到的AI时间节省，是否是真正的时间节省。自从去年秋天撰写《生成式AI与生产率》报告以来（这些研究结果一直反复浮现在我的脑海中：“非营利组织模型评估与威胁研究机构（METR）得到了一项令人担忧的发现。十六名经验丰富的开发者参与了这项研究。他们被分配了246项编码任务，并被要求预测每项任务的完成时间。随后，这些任务被随机分配为允许或不允许使用AI工具。METR的发现是：‘完成研究后，开发者估计，允许使用AI使完成时间缩短了20%。令人意外的是，我们发现允许使用AI实际上使完成时间增加了19%——AI工具拖慢了开发者的速度。’”
要证明目前投入生成式AI的巨额资金具有合理性，全球经济各领域都必须实现生产率的巨大提升。而且，这些提升必须能够衡量，企业才会进行达到改造运营流程、优化AI用例所需水平的投资。正如巴克莱分析师指出的：
“我们的最终结论是，AI采用似乎是渐进而稳定的，而非迅速且具有变革性的；大多数家庭和企业仍表示，其接触程度有限。与此同时，生产率增长出现结构性回升的证据依然出人意料地脆弱……我们在现有的行业层面数据中几乎找不到有说服力的证据，证明更快采用AI的行业已经在经历更强劲的生产率增长。”
面对泡沫化的估值、token成本、芯片通胀、来自中国的竞争，以及越来越多的证据显示企业投入了相当可观的AI支出，却未能实现有意义的投资回报，“有说服力的证据”需要尽快出现，否则近期的市场震荡可能演变为真正的地震。自12月以来，我一直向客户提供对冲这一风险的思路，而这些思路今年迄今的表现显著优于AI交易。正如我在报告中详细阐述的，我并未低估AI的长期潜力、市场自满情绪的持续性，或目前已嵌入市场结构的支持AI的偏向。但我也认为，跳出“卖铲人”和超大规模云服务商的框架思考，可以捕捉到更多alpha。 #资讯#


## 研究交付要求

这是一条已经从整场材料语义拆出的独立催化，不要重新把它压回大盘总评。核实美股异动原因和传播时序；用结构化供应链、内部 KB 与公司披露坐实每个亚洲候选；逐腿查真实行情和本地市场 price-in；候选不成立也要逐条给出有证据的 AVOID。