Lire cette page en français

Agents

Eclipse Enclave sandboxes coding agents, OVHcloud launches 55 AI products with B300 and Rubin GPUs, OpenAI shows its Codex-run Performance Factory; Claude Code 2.1.329 and Codex 0.171.0

The Eclipse Foundation releases Enclave 1.0, an MIT-licensed sandbox that runs Claude Code, Codex or OpenCode in isolated containers; Kong launches Volcano; the GitHub Copilot app reaches Azure DevOps repos. OpenAI says Codex agents ran more than 1,700 performance investigations in about three months, and ships GPT-Live-1.1 in its API. OVHcloud unveils 55 products and up to 60,000 B300 GPUs, Tencent publishes the Hy4 weights, and Claude Code 2.1.329, Codex 0.171.0 and Zed 1.29 ship.

By ·

On November 19 and 20, 2026, the news for developers was less about new agents than about where they run: in a sandbox, on an agent cloud, on Azure DevOps repositories or on European GPUs. Here is what shipped. Unless stated otherwise, figures are the vendors’ own.

Eclipse Enclave 1.0: a vendor-neutral sandbox for coding agents

The Eclipse Foundation released Eclipse Enclave 1.0 on November 19 (Eclipse Foundation). It is an open source runtime and sandbox that runs each agent session in an isolated container, with defined access to files, networks and development resources. It supports Claude Code, Codex, OpenCode and other agents. According to the announcement:

Kong launches Volcano, an agent cloud that fixes its own bugs

Kong announced Volcano on November 19 (Volcano). The platform hosts applications, agents and data storage, and can be driven from Cursor, Codex, Claude Code or an online builder. Once the code runs there, Volcano “continuously” looks for issues and proposes fixes, “with no human intervention” according to Kong. The announcement gives no prices.

GitHub: the Copilot app works on Azure DevOps repositories

Microsoft opened a public preview of the Azure DevOps integration in the GitHub Copilot app on November 19 (Azure DevOps Blog). Agents can now create branches, change code, commit and open pull requests on Azure Repos repositories, and Agent Merge can follow a pull request, address review comments, fix CI failures and merge it when it is ready. The integration is switched on under Settings > Experimental; work items and Projects support are planned.

The same day, GitHub updated the new pull request Overview page, in public preview since November 5: failed optional checks no longer make a mergeable pull request look blocked, and comments that mention Copilot now show the action and the selected model before you submit (GitHub). Red Hat Enterprise Linux 9 and 10 images for larger runners are now generally available on GitHub Team and Enterprise Cloud (GitHub).

OpenAI: Codex agents that chase ChatGPT slowdowns, and GPT-Live-1.1

OpenAI described its Performance Factory on November 19 (OpenAI). When a performance alert fires, its Regression Workflow opens a task; a Codex agent investigates, records its findings and proposes a fix that an engineer reviews; after deployment, another agent checks that performance is back to baseline. In just over three months, the workflow tracked more than 1,700 investigations and found the root cause in more than 600 cases, about 17 automated investigations a day. Engineers still review and merge the fixes; OpenAI says it is exploring whether agents can do that too.

In its API, OpenAI released GPT-Live-1.1 on November 19, a full-duplex voice model that transcribes more accurately and keeps a more consistent accent across languages than GPT-Live-1 (API changelog). It is billed $0.05 per active minute and $0.02 per minute on standby (model page). Standby, available for GPT-Live-1 and 1.1 over WebRTC, pauses inference after 5 to 60 seconds of silence and resumes when someone speaks (guide).

For agent builders, OpenAI also published a guide on November 19: with GPT-6.2, the Responses API drops the reasoning items that follow any edit to earlier history, so it recommends an append-only pattern; a new input_transformations field shows which reasoning items were removed and why (OpenAI guide, OpenAI Developers on X).

OVHcloud: 55 products, B300 and Rubin GPUs

At its AI Summit on November 19, OVHcloud unveiled 55 new cloud and AI products (OVHcloud):

Everything is offered in Public Cloud, single-tenant Private Cloud and On-Prem, including with SecNumCloud qualification. The release gives no prices.

Tencent publishes the Hy4 weights

Tencent released the weights of Hy4, Hy4-FP8 and the pretrained Hy4 base on Hugging Face, ModelScope and GitCode on November 20 (Guandian, model card). Hy4 is a mixture-of-experts model with 770B total parameters and 49B active per token that understands text, images and video. The license is the Tencent Hy License, not Apache 2.0: a company that offers the model as a service needs a separate written license if its worldwide revenue exceeds US$50 million over 12 months, or for a product with more than 100 million monthly users; internal use is exempt (license).

New versions

What we updated

Sources

Spotted an error in this article? Report it · Our corrections policy

OrigoTech logo

New: the app iPhone & iPad

OrigoTech is now an iPhone and iPad app

Read the articles natively, with “New” markers on unread ones, local search and offline reading of what you have already loaded. Free, no account, iOS 17 or later; ads appear between articles.

Rekha icon: a house drawn in a single stroke

From the same author Self-promotion

An Eulerian path, drawn with your finger

Rekha, the iPhone game by OrigoTech's author: connect every dot in a single stroke without tracing the same link twice, across 500 drawings that are all solvable.

The OrigoTech newsletter

A sourced weekly digest of AI tool news.

Read next

Agents

GitHub at Ignite: Copilot Evals, “Adaptive” code review and Copilot in the browser; Foundry adds Baseten; TensorWave commits 457 MW to OpenAI

At Ignite, GitHub puts Copilot Evals in public preview to test your skills, agents and MCP tools, lets Copilot pick the effort for each code review and lets it drive Chrome. Microsoft Foundry adds Baseten and deploys models and agents onto your own AKS clusters. TensorWave will supply 457 MW of AMD capacity to OpenAI. Also: Claude Code 2.1.327, Codex CLI 0.170.0, OpenCode 2.3.2.

· 7 min read

Agents

Antigravity Dynamic Workflows: /run-workflow turns agent teams into Python scripts; Copilot drops six models on December 14, Codex CLI 0.169.0

Antigravity 2.32.0 (Nov 13) adds Dynamic Workflows, where the agent writes a short Python script that runs many subagents in parallel or in sequence, and you approve the script before it runs. GitHub put Gemini 4 Hydrogen in Copilot and will retire six Copilot models on December 14. Also: Codex CLI 0.169.0, Claude Code 2.1.325, Zed 1.28 and AirTrunk's 1.8 GW plan in India.

· 6 min read

Models

Gemini 4 Hydrogen: price, benchmarks and Antigravity's new Orchestrator; Claude Code adds Code Mode, Codex CLI 0.168.0

Google released Gemini 4 Hydrogen on November 12, a fast model for everyday coding and agents at $1.50/$7.50 per million tokens, with a free API tier and a new Orchestrator in Antigravity that pairs it with Carbon or Argon. Also: Code Mode in Claude Code 2.1.323, Codex CLI 0.168.0, the agent-built AWS Emulator and Nebius's $40-45 billion capex plan for 2027.

· 5 min read