Provider announcements · WATCHED

The AI model lifecycle newsroom.

Provider announcements, deprecations, access restrictions and weak signals before they turn into production incidents.

Signal discipline

Not every signal is lifecycle truth.

Zombify separates official provider lifecycle data from weaker availability and news signals. Suspensions, routing changes and aggregator removals deserve review before they become customer-facing alerts.

Zombify tracks public model and platform signals from third-party sources. Zombify is independent and is not affiliated with, endorsed by, or sponsored by the companies or publishers referenced. Source links are provided for attribution and verification.

Featured signal TechCrunch

Source published 09 July 2026

OpenAI is shutting down Atlas, but its AI browser ambitions are still growing

Retirement watch Third-party discovery signal

Possible retirement or shutdown signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Migration risk starts before the shutdown date. The useful work is checking hardcoded IDs, aliases and fallback coverage while there is still room to move.
Read at source: TechCrunch AI →
TechCrunch Source published 24 July 2026

Anthropic launches Opus 5

New model signal Third-party discovery signal

Possible new-model or platform-availability signal. The exact callable model ID still needs review. This is third-party discovery evidence, not lifecycle truth.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
TechCrunch Source published 24 July 2026

As US weighs response to Chinese AI, industry urges against broad open-weight restrictions

Restriction signal Third-party discovery signal mistral

Possible access or policy-risk signal. Detected model tags: mistral. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. Model tags found: mistral. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
Ars Technica Source published 21 July 2026

Google announces Gemini 3.6 Flash and cybersecurity AI, teases 3.5 Pro and Gemini 4

Availability signal Third-party discovery signal gemini-3.6-flash-and-cybersecurity-aigemini-4-there-are-new-3.6-and-3.5-models-todaygemini-4.

Possible availability signal. Detected model tags: gemini-3.6-flash-and-cybersecurity-ai, gemini-4-there-are-new-3.6-and-3.5-models-today and gemini-4.. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: Ars Technica AI →
TechCrunch Source published 14 July 2026

OpenAI’s new flagship model deletes files on its own, people keep warning

Availability signal Third-party discovery signal gpt-5.6

Possible availability signal. Detected model tags: gpt-5.6. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: TechCrunch AI →
TechCrunch Source published 10 July 2026

OpenAI says GPT 5.6 is the ‘preferred model’ for Microsoft Copilot 365 amid breakup chatter

Availability signal Third-party discovery signal gpt-5.6

Possible availability signal. Detected model tags: gpt-5.6. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: TechCrunch AI →
TechCrunch Source published 09 July 2026

OpenAI launches its new family of models with GPT-5.6

New model signal Third-party discovery signal gpt-5.6

Lucas Ropek reports new model activity involving gpt-5.6. Confirm the callable IDs and availability surface before using this as replacement or registry guidance.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: gpt-5.6. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
TechCrunch Source published 09 July 2026

Elon Musk praises Mythos/Fable, promises not to ‘cut off’ Anthropic

Restriction signal Third-party discovery signal

Julie Bort reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
OpenAI Source published 09 July 2026

GPT-5.6 is now the preferred model in Microsoft 365 Copilot

Availability signal Official provider evidence gpt-5.6

Possible availability signal. Detected model tags: gpt-5.6. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
OpenAI Source published 09 July 2026

GPT-5.6: Frontier intelligence that scales with your ambition

API change Official provider evidence gpt-5.6

Possible API, routing, quota or platform-change signal. Detected model tags: gpt-5.6. This is official-provider evidence and can inform registry review.

Zombify take: API and routing changes can break production without changing a model name. Check endpoints, quotas, parameters and provider surfaces before assuming watched stacks are safe.
Read at source: OpenAI news →
TechCrunch Source published 08 July 2026

SpaceXAI releases Grok 4.5, which Elon describes as an ‘Opus-class model’

New model signal Third-party discovery signal grok-4.5

Lucas Ropek reports new model activity involving grok-4.5. Confirm the callable IDs and availability surface before using this as replacement or registry guidance.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: grok-4.5. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
OpenAI Source published 08 July 2026

Introducing GPT-Live

New model signal Official provider evidence

Possible new-model or platform-availability signal. The exact callable model ID still needs review. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenAI news →
TechCrunch Source published 07 July 2026

Discord admits AI moderation bug wrongfully banned users over harmless images

Restriction signal Third-party discovery signal

Lauren Forristal reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
The Decoder Source published 07 July 2026

Deepseek is designing its own AI chip

New model signal Third-party discovery signal

Possible new-model or platform-availability signal. The exact callable model ID still needs review. This is third-party discovery evidence, not lifecycle truth.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: The Decoder AI coverage →
TechCrunch Source published 04 July 2026

What is Mistral AI? Everything to know about the OpenAI competitor

Availability signal Third-party discovery signal mistral

Possible availability signal. Detected model tags: mistral. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: TechCrunch AI →
TechCrunch Source published 01 July 2026

Cloudflare’s new policy pushes AI companies to pay for publishers’ content

Restriction signal Third-party discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
Associated Press Source published 01 July 2026

Trump administration lifts restrictions on Anthropic's Claude models after cybersecurity alarm

Restriction signal Aggregated discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: AP News via AP AI coverage via Google News →
TechCrunch Source published 01 July 2026

Gemini Spark, Google’s agentic assistant, is now available on Mac

Availability signal Third-party discovery signal

Possible availability signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: TechCrunch AI →
TechCrunch Source published 01 July 2026

Trump drops restrictions on Anthropic’s Mythos and Fable models

Restriction signal Third-party discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
TechCrunch Source published 30 June 2026

Anthropic launches Claude Sonnet 5 as a cheaper way to run agents

New model signal Third-party discovery signal claude-sonnet-5gpt-5.5

Rebecca Bellan reports new model activity involving claude-sonnet-5 and gpt-5.5. Confirm the callable IDs and availability surface before using this as replacement or registry guidance.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-sonnet-5 and gpt-5.5. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
TechCrunch Source published 30 June 2026

Anthropic’s Claude Science bets on workflow, not a new model, to win over scientists

New model signal Third-party discovery signal

Possible new-model or platform-availability signal. The exact callable model ID still needs review. This is third-party discovery evidence, not lifecycle truth.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
OpenAI Source published 30 June 2026

How ChatGPT adoption has expanded

Availability signal Official provider evidence

Possible availability signal. No callable model ID has been confirmed yet. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
Anthropic Source published 30 June 2026

Claude Fable 5 access restored

Availability restored Official provider evidence claude-fable-5

Anthropic says export controls on Fable 5 and Mythos 5 have been lifted, and Claude Fable 5 is being restored for global users across Claude Platform, Claude.ai, Claude Code and Claude Cowork.

Zombify take: Move Claude Fable 5 back to active once your own account or provider route confirms access. Keep Mythos 5 separate because it remains a controlled-access model, not a general replacement path.
Read at source: Anthropic update →
Business Insider Source published 27 June 2026

Anthropic's Mythos 5 gets a limited carveout from US restrictions

Restriction signal Aggregated discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: Business Insider →
TechCrunch Source published 27 June 2026

Asian AI startups launch Mythos-like models as Anthropic’s export ban drags on

Restriction signal Third-party discovery signal

Kate Park reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
TechCrunch Source published 26 June 2026

OpenAI limits GPT-5.6 rollout after government request, says restrictions shouldn’t be the norm

Restriction signal Third-party discovery signal gpt-5.6

Possible access or policy-risk signal. Detected model tags: gpt-5.6. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. Model tags found: gpt-5.6. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
Business Insider Source published 26 June 2026

OpenAI says access to its new GPT-5.6 model is limited at the US government's request

Availability signal Aggregated discovery signal gpt-5.6

Possible availability signal. Detected model tags: gpt-5.6. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: Business Insider →
OpenAI Source published 26 June 2026

Previewing GPT-5.6 Sol: a next-generation model

New model signal Official provider evidence gpt-5.6

Possible availability signal. Detected model tags: gpt-5.6. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: gpt-5.6. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenAI news →
TechCrunch Source published 25 June 2026

The White House is asking OpenAI to slow roll the release of its new model over safety concerns

New model signal Third-party discovery signal gpt-5.6

Lucas Ropek reports new model activity involving gpt-5.6. Confirm the callable IDs and availability surface before using this as replacement or registry guidance.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: gpt-5.6. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: TechCrunch AI →
Business Insider Source published 24 June 2026

Google delays Gemini 3.5 Pro launch to July as it tweaks its new frontier AI model

New model signal Aggregated discovery signal gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model-business-insider-google-delays-gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model

Business Insider reports new model activity involving gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model-business-insider-google-delays-gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model. Confirm the callable IDs and availability surface before using this as replacement or registry guidance.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model-business-insider-google-delays-gemini-3.5-pro-launch-to-july-as-it-tweaks-its-new-frontier-ai-model. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: Business Insider →
OpenAI Source published 23 June 2026

How GPT-5 helped immunologist Derya Unutmaz solve a 3-year-old mystery

Availability signal Official provider evidence gpt-5

Possible availability signal. Detected model tags: gpt-5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
TechCrunch Source published 19 June 2026

Is the US government’s Anthropic ban accidentally helping the brand?

Restriction signal Third-party discovery signal

Theresa Loconsolo reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
OpenAI Source published 18 June 2026

Improving health intelligence in ChatGPT

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
TechCrunch Source published 17 June 2026

World leaders want American AI. They just don’t want America to be able to turn it off.

Restriction signal Third-party discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
TechCrunch Source published 17 June 2026

Anthropic becomes first AI startup to join the Frontier carbon removal coalition

Retirement watch Third-party discovery signal

Possible retirement or shutdown signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Migration risk starts before the shutdown date. The useful work is checking hardcoded IDs, aliases and fallback coverage while there is still room to move.
Read at source: TechCrunch AI →
OpenAI Source published 17 June 2026

A near-autonomous AI chemist improves a challenging reaction in medicinal chemistry

Availability signal Official provider evidence gpt-5.4

Possible availability signal. Detected model tags: gpt-5.4. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
TechCrunch Source published 15 June 2026

The US government’s Anthropic models ban was never about an AI jailbreak

Restriction signal Third-party discovery signal

Zack Whittaker reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
Business Insider Source published 15 June 2026

Anthropic's new models were restricted by the US. Europe's top AI startup has been waiting for this moment.

Restriction signal Aggregated discovery signal

Business Insider reports a possible access restriction involving Business Insider. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: Business Insider →
TechCrunch Source published 15 June 2026

Cybersecurity vets protest ‘dangerous’ US government ban on Anthropic’s most powerful models

Restriction signal Third-party discovery signal

Lorenzo Franceschi-Bicchierai reports a possible access restriction involving TechCrunch. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
OpenRouter Source published 15 June 2026

How to Use Hermes Agent with OpenRouter: Setup, Models & Routing Hermes Agent has pushed over 17 trillion tokens through OpenRouter. Here's how to set it up,…

Restriction signal Official provider evidence claude-fable-5

Possible access or policy-risk signal. Detected model tags: claude-fable-5. This is official-provider evidence and can inform registry review.

Zombify take: This is a watchlist item, not a verdict. Model tags found: claude-fable-5. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: OpenRouter announcements →
Business Insider Source published 14 June 2026

Inside the whirlwind 24 hours that led the White House to slap export controls on Anthropic

Restriction signal Aggregated discovery signal

Business Insider reports that White House officials moved to restrict foreign access to Anthropic’s Mythos and Fable models after security concerns were raised around model safeguards. Anthropic reportedly disputed parts of the account, but the access impact is material enough to treat as a live availability and policy-risk signal.

Zombify take: This is the kind of signal that can break assumptions fast: a model can move from newly available to restricted because of policy, safety or national-security pressure. Treat Mythos and Fable as watched IDs until Anthropic’s official access position, affected customer scope and replacement path are clear.
Read at source: Business Insider via Business Insider AI coverage →
TechCrunch Source published 13 June 2026

Amazon CEO reportedly raised Anthropic model concerns before government crackdown

Restriction signal Third-party discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
Business Insider Source published 13 June 2026

Tech world reacts to Trump administration ban on foreign access to Anthropic's new AI models

Restriction signal Aggregated discovery signal

Business Insider reports a possible access restriction involving Business Insider. No callable model IDs have been confirmed from provider evidence yet. Verify against official provider evidence before changing registry guidance.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: Business Insider →
TechCrunch Source published 13 June 2026

As Anthropic suspends access to new models, India debates its AI future

Restriction signal Third-party discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: TechCrunch AI →
OpenRouter Source published 12 June 2026

Classifiers: Track What Your Agents Do and What It Costs Define a taxonomy and a small model tags every generation in your workspace by department, task type,…

API change Official provider evidence claude-opus-4.8gpt-5.5

Possible API, routing, quota or platform-change signal. Detected model tags: claude-opus-4.8 and gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: API and routing changes can break production without changing a model name. Check endpoints, quotas, parameters and provider surfaces before assuming watched stacks are safe.
Read at source: OpenRouter announcements →
Business Insider Source published 09 June 2026

Anthropic releases Claude Fable 5, a 'Mythos-class' AI model with safeguards

Availability signal Aggregated discovery signal claude-fable-5

Possible availability signal. Detected model tags: claude-fable-5. This is third-party discovery evidence, not lifecycle truth.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: Business Insider via Business Insider AI coverage →
OpenAI Source published 09 June 2026

How engineers at Nextdoor use Codex to build without limits

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
Business Insider Source published 08 June 2026

OpenAI, Anthropic, and Nvidia ramped up H-1B filings before a judge blocked Trump's $100,000 fee

Restriction signal Aggregated discovery signal

Possible access or policy-risk signal. No callable model ID has been confirmed yet. This is third-party discovery evidence, not lifecycle truth.

Zombify take: This is a watchlist item, not a verdict. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: Business Insider via Business Insider AI coverage →
Ars Technica Source published 03 June 2026

Google's new Gemma 4 12B model is designed to run on any laptop with 16GB of RAM

API change Third-party discovery signal gemma-4

Ryan Whitwam reports a possible API or platform-surface change involving Ars Technica. Detected model tags: gemma-4. Check affected endpoints, routing, quotas and customer stacks before publishing guidance.

Zombify take: API and routing changes can break production without changing a model name. Check endpoints, quotas, parameters and provider surfaces before assuming watched stacks are safe.
Read at source: Ars Technica AI →
OpenRouter Source published 01 June 2026

Choosing the Optimal Image Input Detail Level in LLMs We ran 1,730 visual reasoning questions across 5 models. Dropping image detail to "low" costs real…

New model signal Official provider evidence claude-opus-4.8.gpt-5.5gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8., gpt-5.5 and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8., gpt-5.5 and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

The OpenRouter MCP Server Connect your coding agent to OpenRouter's live model catalog, benchmarks, docs, and test inference, all without leaving your editor.…

New model signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

The Open Weight Models that Matter: June 2026 A slew of compelling open-weight models have shipped from new players in both China and the US. As of June 2026,…

New model signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

DeepSeek V4 Is Earning Agentic Token Share DeepSeek doubled its token share on OpenRouter in six months. V4 Flash is the model that made it happen, and…

New model signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

Introducing the Unified Image API A dedicated Image API with capability discovery across 30+ models from 8 providers. One endpoint tells your code what each…

Restriction signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible access or policy-risk signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: This is a watchlist item, not a verdict. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

AI Governance Checklist: Your LLM Architecture Comes First Policy language can't show who called which model or where the audit trail lives. Map your…

Restriction signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible access or policy-risk signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: This is a watchlist item, not a verdict. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. The risk is whether model access, policy, routing or regional availability is shifting underneath production stacks.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

OpenRouter vs Portkey: Which LLM Gateway for Your Team? OpenRouter routes across providers on credits you buy; Portkey governs the provider keys you already…

New model signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 01 June 2026

Kilo Code + OpenRouter: Setup, Model Routing, and Free Models Kilo Code is a bring-your-own-provider coding agent, so adding OpenRouter gives it one key for…

New model signal Official provider evidence claude-opus-4.8.gemini-3.5-flash-and-claude-opus-4.8.-june-1

Possible availability signal. Detected model tags: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. This is official-provider evidence and can inform registry review.

Zombify take: New-model coverage is useful only when it maps to a real callable ID. Model tags found: claude-opus-4.8. and gemini-3.5-flash-and-claude-opus-4.8.-june-1. Until the API surface and availability are clear, it belongs in review rather than replacement guidance.
Read at source: OpenRouter announcements →
OpenRouter Source published 28 May 2026

Surpassing Frontier Performance with Fusion A panel of budget models, fused through OpenRouter, outscored GPT-5.5 and Claude Opus 4.8 on 100 complex research…

API change Official provider evidence claude-opus-4.8gpt-5.5

Possible API, routing, quota or platform-change signal. Detected model tags: claude-opus-4.8 and gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: API and routing changes can break production without changing a model name. Check endpoints, quotas, parameters and provider surfaces before assuming watched stacks are safe.
Read at source: OpenRouter announcements →
OpenAI Source published 27 May 2026

Warp’s big bet on building open source with GPT-5.5

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
OpenAI Source published 15 May 2026

Databricks brings GPT-5.5 to enterprise agent workflows

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
OpenAI Source published 07 May 2026

Scaling Trusted Access for Cyber with GPT-5.5 and GPT-5.5-Cyber

Availability signal Official provider evidence gpt-5.5gpt-5.5-cyber

Possible availability signal. Detected model tags: gpt-5.5, gpt-5.5-cyber. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
OpenAI Source published 05 May 2026

GPT-5.5 Instant: smarter, clearer, and more personalized

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →
OpenAI Source published 05 May 2026

GPT-5.5 Instant System Card

Availability signal Official provider evidence gpt-5.5

Possible availability signal. Detected model tags: gpt-5.5. This is official-provider evidence and can inform registry review.

Zombify take: Availability signals are smoke, not proof. Use this to look for changes in callable models, routes, quotas, regions or access policy before it becomes customer-visible risk.
Read at source: OpenAI news →

Newsroom cards are signal indexes with Zombify commentary, not article republication. Corrections or source concerns: hello@zombify.au.