# GDPx Benchmark: Claude Opus 5.5 Leads at Just 12%

> SemiAnalysis's GDPx tests agents on 206 real-world tasks in 11 domains. Claude Opus 5.5 leads at 12%; finance and legal score zero.

- Author: Nitish Garg, Founder & CEO, CellCog
- Published: 2026-11-02
- Canonical (HTML): https://cellcog.ai/blog/gdpx-benchmark/
- Section: Guides / Choosing a platform
- Publisher: CellCog (https://cellcog.ai), the AI employee platform. Blog index for agents: https://cellcog.ai/blog/llms.txt

## Key points

- SemiAnalysis released GDPx on November 2, 2026: 206 agent tasks across 11 high-value domains, built with 14 environment startups.
- Claude Opus 5.5 leads at 12.0%, ahead of Gemini 4 Carbon at 10.9% and GPT-6 Astra at 9.6%.
- Each task drops the agent into a working environment, such as a deal room of about 2,000 files, with up to 8 hours to finish.
- Finance and legal tasks scored zero for nearly every model.
- SemiAnalysis says safety classifiers held back Opus 5.5 and GPT-6 Astra in cybersecurity, biology and alignment.

## At a glance

- **What is GDPx?** A SemiAnalysis benchmark of 206 agent tasks across 11 domains, each set inside a realistic work environment, released November 2, 2026.
- **Which model leads?** Claude Opus 5.5 at 12.0%, then Gemini 4 Carbon at 10.9% and GPT-6 Astra at 9.6%.
- **Why so low?** The tasks are built to be hard for frontier models, finance and legal scored near zero, and safety classifiers cut some scores.

SemiAnalysis released [GDPx](https://gdpx.semianalysis.com/) on November 2, 2026, a benchmark of 206 agent tasks across 11 domains that pay, from chip design to quant trading. Its [launch thread](https://x.com/SemiAnalysis_/status/2117075900575477855) at 02:06 UTC put the result in one line: "GDPx is a true frontier-level benchmark: even the best model scores just 12%" and passed a million views within a day.

The best model is Claude Opus 5.5 at 12.0%, ahead of Gemini 4 Carbon at 10.9% and GPT-6 Astra at 9.6%. Finance and legal tasks scored zero for almost every model. The difference from older benchmarks is the setting: each task drops the agent into a working environment, such as a deal room of about 2,000 files, and leaves it to find what it needs.

## What GDPx tests

![Animated diagram: on the left a prompt card and one PDF feed straight into a model chip; on the right an agent walks into a grid of deal-room folders, finds the one amber folder and only then produces its answer, under a 12% top score](https://cellcog.ai/blog/media/gdpx-benchmark/env.svg)
*GDPval hands the model a prompt; GDPx makes the agent find the context*

SemiAnalysis built the tasks with 14 startups that make training environments for AI labs, one or more per domain, and ran each model in its own agent harness. The method, per the [SemiAnalysis write-up](https://newsletter.semianalysis.com/p/gdpx-measuring-frontier-model-capabilities):

*Table: How GDPx v1.0 is run, per SemiAnalysis (read November 2, 2026)*

| Item | GDPx v1.0 |
|---|---|
| Tasks | 206 |
| Domains | 11: chip design, cybersecurity, biology, finance, legal, quant trading, recursive self-improvement, hardware design, materials science, alignment, software engineering |
| Data partners | 14 environment startups, at least one per domain |
| Harness | Each model's own: Claude Code for Opus 5.5, Codex for GPT-6 Astra, Antigravity for Gemini 4 Carbon; OpenCode otherwise |
| Time limit | 8 hours by default |
| Score | pass@1 across three attempts; every domain weighted equally |

The finance example shows what "environment" means. The agent works a private-equity deal from the seat of one adviser, with the target company's data room and that adviser's deal folder loaded: about 4,000 anonymized files in all. A human associate did the same task on the real transaction.

## The leaderboard

The public site shows 13 of the 20 models it tested; these are its scores.

*Table: GDPx v1.0 overall scores, per the GDPx leaderboard (read November 2, 2026)*

| Rank | Model | GDPx score |
|---|---|---|
| 1 | Claude Opus 5.5 | 12.0% |
| 2 | Gemini 4 Carbon | 10.9% |
| 3 | GPT-6 Astra | 9.6% |
| 4 | Muse Wonder Full | 9.0% |
| 5 | Claude Fable 5.1 | 8.5% |
| 6 | GPT 6.1 Sol | 8.4% |
| 7 | Claude Sonnet 5.5 | 8.2% |
| 8 | Gemini 4 Argon | 7.3% |
| 9 | Muse Wonder | 6.8% |
| 10 | DeepSeek 4.1 Flash 1017 | 5.4% |
| 11 | GPT-6 Luna | 5.1% |
| 12 | Grok 4.7 | 3.9% |
| 13 | GLM 5.3 | 2.7% |

On cost, SemiAnalysis reads its own chart this way: "Luna and DeepSeek fight for the low end, Opus wins the high end, and OpenAI claims the middle." Gemini 4 Carbon, not yet generally available, sits second overall and is the model the authors say to watch.

## Who wins where

The overall number hides sharp spikes. In SemiAnalysis's words: "Opus 5.5 is the best all-around model and wins in quant trading, hardware design, and chip design."

*Table: Domain winners in GDPx v1.0, per SemiAnalysis (November 2, 2026)*

| Domain | Winner |
|---|---|
| Chip design | Claude Opus 5.5 |
| Hardware design | Claude Opus 5.5 |
| Quant trading | Claude Opus 5.5 |
| Biology | Gemini 4 Carbon |
| Software engineering | Gemini 4 Carbon |
| Recursive self-improvement | GPT-6 Astra |
| Materials science | GPT-6 Astra |
| Cybersecurity | Muse Wonder Full |
| Finance and legal | Nearly every model scores 0 |

*GDPx in three pictures: the top five, who wins where, and what makes it hard*

![Flat bar panel of the GDPx top five: Claude Opus 5.5 12.0% in amber, Gemini 4 Carbon 10.9%, GPT-6 Astra 9.6%, Muse Wonder Full 9.0%, Claude Fable 5.1 8.5% in teal](https://cellcog.ai/blog/media/gdpx-benchmark/slide1.webp)
*The top five: Claude Opus 5.5 at 12.0%, then Gemini 4 Carbon, GPT-6 Astra and Muse Wonder*

![Flat panel with three columns: Claude Opus 5.5 over chip design, hardware design and quant trading; Gemini 4 Carbon over biology and software engineering; GPT-6 Astra over self-improvement and materials science](https://cellcog.ai/blog/media/gdpx-benchmark/slide2.webp)
*Who wins where: Opus in chips and trading, Carbon in bio and code, Astra in research*

![Flat panel: on the left GDPval, a prompt card and one page feeding a chip; on the right GDPx, a walking chip heading into a grid of folders with one amber folder, labelled a 2,000-file deal room](https://cellcog.ai/blog/media/gdpx-benchmark/slide3.webp)
*Spoon-fed vs found: GDPval gives a prompt and one PDF, GDPx a 2,000-file deal room*

Domain wins are not uniform inside a domain either. On one chip-design task from the partner Ergodex, timing fixes on a real 20,702-cell design, GPT-6 Astra averaged 0.34 against 0.135 for Claude Opus and 0.07 for Fable.

## Why the scores are so low

Three reasons come out of the write-up. The first is design: SemiAnalysis set out to make tasks the best models solve less than a fifth of the time, and says its review process, run on Terminal Bench 4.0, "flags nearly 50% of tasks as broken." Partners had to pass that review on the first try.

The second is safety classifiers. The authors write that "certain models like Opus 5.5 and GPT-6 Astra have sandbagged overall scores due to their safety classifiers", which can refuse work in cybersecurity, biology and alignment. Remove those three domains and the ranking changes; that is also why "Muse Wonder Full is actually the best available cyber model today by a meaningful margin" in their reading.

The third is finance and legal: "Our finance and legal tasks are so challenging that basically every model scores exactly 0."

GDPx is pitched as the successor to OpenAI's GDPval, released September 25, 2025. SemiAnalysis's complaint is that "Every GDPval task is precisely specified and non-interactive", with the context handed over in the prompt, and that "Human expert inter-rater agreement was just 71% on the gold subset". Ethan Mollick [called it](https://x.com/emollick/status/2117143076443746659) a useful effort and noted that "benchmarks of expert-level work are incredibly hard to pull off."

## What this means if your AI employees run on Claude

Our conflict, declared: we build [CellCog](https://cellcog.ai/ai-employees), and every tier of our AI employees runs on Claude Opus 5.5, the model at the top of this board, so a good result for it is good for us. Read our take with that in mind.

The useful part for anyone hiring an agent is the shape of the test, not the rank. GDPx tasks look like real work: hours long, with the context buried in folders the agent has to search. A 12% top score says the hardest expert work still mostly fails. It does not say that routine work fails, because GDPx is built to sit at the frontier on purpose. Our AI employees take on everyday business work with an owner approving anything that leaves the company, and that is a different bar from an eight-hour chip-timing task.

## What we are watching

- **Gemini 4 Carbon going generally available.** It is second overall and first in biology and software engineering on a model most people cannot use yet.
- **The next version.** SemiAnalysis says "We plan to release new versions every 3-6 months" and calls GDPx a living benchmark.
- **Labs reporting it.** Whether GDPx joins GDPval on model release cards.

## Sources

- SemiAnalysis: [GDPx write-up](https://newsletter.semianalysis.com/p/gdpx-measuring-frontier-model-capabilities) by Max Kan, Andrew Megalaa and Dylan Patel, and the [GDPx leaderboard](https://gdpx.semianalysis.com/), read November 2, 2026.
- X: [@SemiAnalysis_ launch thread](https://x.com/SemiAnalysis_/status/2117075900575477855) and [@emollick](https://x.com/emollick/status/2117143076443746659); times from the post ids.

## FAQ

**Who made GDPx?**

SemiAnalysis, with Max Kan, Andrew Megalaa and Dylan Patel as authors and 14 data partners that build training environments.

**How is GDPx scored?**

pass@1 across three attempts, with all 11 domains weighted equally and an 8-hour default time limit, each model in its own harness.

**How is GDPx different from GDPval?**

GDPval gives the model its context in the prompt; GDPx makes the agent find it in a working environment, such as a deal room of thousands of files.

**Which model is best at cybersecurity on GDPx?**

SemiAnalysis names Muse Wonder Full, and notes that safety classifiers hold back Opus 5.5 and GPT-6 Astra in that domain.

**Will GDPx be updated?**

Yes. SemiAnalysis plans a new version every three to six months.

## Related

- [Intelligence Index v5: Claude Opus 5.5 Leads at 58](https://cellcog.ai/blog/artificial-analysis-intelligence-index-v5/index.md)
- [Every Agent Tier Now Runs on Claude Opus 5.5, Day One](https://cellcog.ai/blog/cellcog-opus-5-5-day-one/index.md)

## The AI employee for this read

[AI Research Assistant](https://cellcog.ai/ai-employees/ai-research-assistant): I built this page. For what it covers, hire a research assistant: it reads the primary sources and cites every line.

---

Markdown alternate of https://cellcog.ai/blog/gdpx-benchmark/. Try CellCog free, no credit card needed: https://cellcog.ai/signup