Anthropic reveals new Opus 4.7 model with focus on advanced software engineering
Source: 9to5Mac
Opus 4.7 needs less supervision for harder coding tasks
Claude Opus 4.7 is the latest generally available version of Anthropic’s main AI model, focused on advanced software development. Compared with Opus 4.6, it shows notable gains on the most difficult coding tasks. Users report being able to hand off their hardest work—previously requiring close supervision—to Opus 4.7 with confidence. The model handles complex, long‑running tasks with rigor, pays precise attention to instructions, and verifies its own outputs before reporting back.
Anthropic says the model has both better vision and a more refined “taste” for creating higher‑quality work. It can see images at higher resolution and produces more polished interfaces, slides, and documents.
Benchmarks show Opus 4.7 ahead of Opus 4.6, GPT‑5.4, and Gemini 3.1 Pro, though it remains behind the more broadly capable Claude Mythos Preview (which is not generally available). See the benchmark comparison table in Anthropic’s blog post here.
Anthropic highlights improvements in instruction following, multimodal support, real‑world work, and memory. As the company notes:
“Opus 4.7 is better at using file system‑based memory. It remembers important notes across long, multi‑session work, and uses them to move on to new tasks that, as a result, need less up‑front context.”
Anthropic is shipping Claude Opus upgrades every two months
Anthropic has established a predictable cadence for directly upgrading its Claude Opus model. Opus 4.7 arrives two months after Opus 4.6, which followed a similar two‑month gap after Opus 4.5. The gap between Opus 4.1 and Opus 4.5 was longer.
Token‑usage changes
Opus 4.7 introduces two changes that affect token usage:
- Updated tokenizer – Improves text processing but can map the same input to 1.0–1.35× more tokens, depending on content type.
- Higher effort levels – The model thinks more deeply on later turns in agentic settings, improving reliability on hard problems but generating more output tokens.
Box’s Head of AI Yashodha Bhavnani notes efficiency gains based on the company’s evaluations:
“Claude Opus 4.7 demonstrates significant efficiency gains while preserving the performance of Claude Opus 4.6. In Box’s evaluations, Opus 4.7 had a 56 % reduction in model calls and a 50 % reduction in tool calls. It also responded 24 % faster and used 30 % fewer AI Units – all enhancements that will help enterprises move faster and scale more affordably.”
Anthropic provides a migration guide here and the Claude Opus 4.7 System Card here.
Claude Code enhancements
Anthropic continues iterating on Claude Code, part of the Claude Mac app:
- Anthropic is giving Claude the ability to use your Mac for you
- Anthropic adds routines to redesigned Claude Code, here’s how it works
- Claude Code gives developers ‘auto mode,’ a safer alternative to skipping permissions
Starting today, auto mode in Claude Code is available for Max plan subscribers, not just Teams/Enterprise/API customers.
Claude Code also gains a new /ultrareview command, which “runs a dedicated review session that reads through your changes and flags what a careful reviewer would catch.”