# AI updates: the timeline of model releases, price changes and shutdowns

Updated 19 October 2026. Every event is read from the vendor's own changelog, pricing, model or deprecation page, linked under it. Rates are vendor list rates per 1M tokens — what your provider bills, not an INFRO charge.

HTML: https://infro.io/ai-updates · JSON: https://infro.io/ai-updates.json · RSS: https://infro.io/ai-updates/feed.xml

## Upcoming (32)

### 20 October 2026 — Google Cloud: Gemini 2.5 Pro, Flash and Flash-Lite retire on Vertex AI

Shutdown. Vertex AI retires the three Gemini 2.5 models and lists Gemini 3.8 Flash among the replacements. On the Gemini API, 2.5 is limited to existing users, not deprecated.

Models: [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash), `gemini-2.5-pro`, `gemini-2.5-flash`, `gemini-2.5-flash-lite`

Source: [Google Cloud, Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions), read 2026-10-19. Permalink: https://infro.io/ai-updates#vertex-gemini-2-5-retired

### 22 October 2026 (earliest) — Google: Veo 3.1 preview ids leave the Gemini API

Shutdown. veo-3.1-generate-preview, -fast- and -lite- stop answering on the Gemini API; Google recommends Gemini Omni Flash. Vertex AI keeps its own dates.

Models: [Veo 3.1](https://infro.io/models/google/veo-3.1), [Veo 3.1 Fast](https://infro.io/models/google/veo-3.1-fast), [Veo 3.1 Lite](https://infro.io/models/google/veo-3.1-lite), [Gemini Omni Flash](https://infro.io/models/google/gemini-omni-flash)

Source: [Google, Gemini deprecations](https://ai.google.dev/gemini-api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#veo-3-1-preview-shutdown

### 23 October 2026 — Google: Deep Research Pro preview shuts down

Shutdown. deep-research-pro-preview-12-2025 stops answering; Google recommends deep-research-preview-04-2026 or its -max- variant.

Models: `deep-research-pro-preview-12-2025`

Source: [Google, Release notes](https://ai.google.dev/gemini-api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#deep-research-pro-preview-shutdown

### 23 October 2026 — OpenAI: OpenAI shuts down older snapshots and gpt-image-1

Shutdown. Including gpt-3.5-turbo-0125, gpt-4-turbo, gpt-4.1-nano, gpt-4o-2024-05-13, the dated o1, o3-mini and o4-mini snapshots, and gpt-image-1 (replaced by GPT Image 2.5).

Models: [GPT Image 1](https://infro.io/models/openai/gpt-image-1), [GPT Image 2.5 Sunburst](https://infro.io/models/openai/gpt-image-2.5-sunburst)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#openai-oct-23-shutdowns

### 24 October 2026 — Perplexity: Older OpenAI model ids leave Perplexity's Agent API

Shutdown. At 00:00 UTC, gpt-5.4, its mini and nano, gpt-5.2, gpt-5.1, gpt-5 and gpt-5-mini retire from the Agent and Orchestrator APIs.

Models: [GPT-5.4 nano](https://infro.io/models/openai/gpt-5.4-nano), `gpt-5.4`, `gpt-5.4-mini`, `gpt-5.2`, `gpt-5.1`, `gpt-5`, `gpt-5-mini`

Source: [Perplexity, Changelog](https://docs.perplexity.ai/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#perplexity-gpt-5-ids-retire

### 31 October 2026 — DeepSeek: DeepSeek starts billing server-side tool calls

Price. Server-side tool calls through DeepSeek Harness, Claude Code and Codex, free until now, bill from 16:00 UTC.


Source: [DeepSeek, DeepSeek-V4.1-Flash-1017](https://api-docs.deepseek.com/news/news261017/), read 2026-10-19. Permalink: https://infro.io/ai-updates#deepseek-tool-billing

### 31 October 2026 — Mistral AI: GLM 5.2 retires on Mistral's platform

Shutdown. Mistral replaces GLM 5.2 with GLM 5.3 on its own platform.

Models: [GLM-5.2](https://infro.io/models/zai/glm-5.2), [GLM-5.3](https://infro.io/models/zai/glm-5.3)

Source: [Mistral AI, Changelog](https://docs.mistral.ai/getting-started/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#mistral-glm-5-2-retired

### 31 October 2026 — OpenAI: OpenAI Evals become read-only

Platform. Existing evals turn read-only ahead of the 30 November shutdown of the Evals platform.


Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-19. Permalink: https://infro.io/ai-updates#openai-evals-read-only

### 2 November 2026 — xAI: grok-imagine-image-quality retires

Shutdown. Requests are served by grok-imagine-image-2.0 at quality low, at a lower per-image price, with the same request and response shape.

Models: [Grok Imagine Image Quality](https://infro.io/models/xai/grok-imagine-image-quality), [Grok Imagine 2.0](https://infro.io/models/xai/grok-imagine-image-2)

Source: [xAI, Release notes](https://docs.x.ai/developers/release-notes), read 2026-10-19. Permalink: https://infro.io/ai-updates#grok-imagine-quality-retired

### 11 November 2026 — OpenRouter: Seedance 1.5 Pro expires on OpenRouter

Shutdown. OpenRouter's own expiration date for bytedance/seedance-1-5-pro. It covers OpenRouter only, not the other providers serving the model.

Models: [Seedance 1.5 Pro](https://infro.io/models/bytedance/seedance-1.5-pro)

Source: [OpenRouter, Models API](https://openrouter.ai/api/v1/models?output_modalities=all), read 2026-10-17. Permalink: https://infro.io/ai-updates#seedance-1-5-pro-openrouter

### 17 November 2026 (earliest) — Google: Gemini preview TTS and Live models shut down

Shutdown. Five preview text-to-speech, Live and native-audio ids leave the Gemini API. Google recommends gemini-3.8-flash-tts, -flash-lite-tts or gemini-3.8-live.

Models: [Gemini 3.1 Flash TTS](https://infro.io/models/google/gemini-3.1-flash-tts), `gemini-3.1-flash-live-preview`, `gemini-2.5-flash-native-audio-preview-12-2025`, `gemini-2.5-flash-preview-tts`, `gemini-2.5-pro-preview-tts`

Source: [Google, Gemini deprecations](https://ai.google.dev/gemini-api/docs/deprecations), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-3-1-flash-tts-shutdown

### 19 November 2026 — Microsoft Azure: o-series models retire on Azure OpenAI

Shutdown. o1, o1-pro, o3, o3-pro, o3-mini, o4-mini and o3-deep-research retire. Azure recommends GPT-5.6 Sol or Terra.

Models: [o3](https://infro.io/models/openai/o3), [o3-pro](https://infro.io/models/openai/o3-pro), [GPT-5.6 Sol](https://infro.io/models/openai/gpt-5.6-sol), [GPT-5.6 Terra](https://infro.io/models/openai/gpt-5.6-terra)

Source: [Microsoft, Model retirement schedule](https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule), read 2026-10-19. Permalink: https://infro.io/ai-updates#azure-o-series-retired

### 19 November 2026 — Google Cloud: Gemini 3.6 Flash retires on Vertex AI

Shutdown. Vertex AI names Gemini 3.8 Flash as the replacement.

Models: [Gemini 3.6 Flash](https://infro.io/models/google/gemini-3.6-flash), [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash)

Source: [Google Cloud, Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions), read 2026-10-19. Permalink: https://infro.io/ai-updates#vertex-gemini-3-6-flash-retired

### 21 November 2026 — OpenAI: GPT-5.6 Sol promotional price guaranteed until here

Price. OpenAI guarantees the promotional rate at least through this date. Budgets that run past it should use the standard rate.

Models: [GPT-5.6 Sol](https://infro.io/models/openai/gpt-5.6-sol)

Source: [OpenAI, Pricing](https://developers.openai.com/api/docs/pricing), read 2026-10-06. Permalink: https://infro.io/ai-updates#gpt-5-6-sol-promotion

### 30 November 2026 — Anthropic: Claude Sonnet 4.5 retires

Shutdown. Requests to Sonnet 4.5 stop working. Anthropic's replacement is Sonnet 5.5.

Models: [Claude Sonnet 5.5](https://infro.io/models/anthropic/claude-sonnet-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-sonnet-4-5-retired

### 30 November 2026 — OpenAI: Prompts API, Agent Builder and Evals shut down

Shutdown. OpenAI ends the reusable prompts API, Agent Builder and the Evals platform.


Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#openai-agent-builder-evals-shutdown

### 1 December 2026 — Google: Gemini 4 Argon price doubles

Price. From $2 input and $10 output to $4 and $20 per 1M tokens; output includes thinking tokens.

Models: `gemini-4-argon`

Source: [Google, Gemini Developer API pricing](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-4-argon-price

### 1 December 2026 — OpenAI: gpt-image-1.5 and gpt-image-1-mini shut down

Shutdown. OpenAI recommends GPT Image 2.5 Sunburst or GPT Image 2.5 Flare.

Models: [GPT Image 1.5](https://infro.io/models/openai/gpt-image-1.5), [GPT Image 1 Mini](https://infro.io/models/openai/gpt-image-1-mini), [GPT Image 2.5 Sunburst](https://infro.io/models/openai/gpt-image-2.5-sunburst), [GPT Image 2.5 Flare](https://infro.io/models/openai/gpt-image-2.5-flare)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#gpt-image-1-5-shutdown

### 11 December 2026 — OpenAI: GPT-5 and o3 dated snapshots shut down

Shutdown. The 2025 GPT-5, GPT-5 mini, nano and pro snapshots and o3-2025-04-16 and o3-pro-2025-06-10 stop answering. OpenAI recommends GPT-5.6 Sol, Terra or Luna.

Models: [o3](https://infro.io/models/openai/o3), [o3-pro](https://infro.io/models/openai/o3-pro), [GPT-5.6 Sol](https://infro.io/models/openai/gpt-5.6-sol), [GPT-5.6 Terra](https://infro.io/models/openai/gpt-5.6-terra), [GPT-5.6 Luna](https://infro.io/models/openai/gpt-5.6-luna), `gpt-5-2025-08-07`, `gpt-5-mini-2025-08-07`, `gpt-5-nano-2025-08-07`, `gpt-5-pro-2025-10-06`

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-19. Permalink: https://infro.io/ai-updates#o3-snapshots-shutdown

### 15 December 2026 — Microsoft Azure: whisper, tts and tts-hd retire on Azure OpenAI

Shutdown. Azure retires whisper 001, tts 001 and tts-hd 001 and lists no replacement.


Source: [Microsoft, Model retirement schedule](https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule), read 2026-10-19. Permalink: https://infro.io/ai-updates#azure-whisper-tts-retired

### 1 January 2027 — Google: Gemini 3.8, 3.7 and 3.6 Flash prices double

Price. From $0.75 input and $3.75 output to $1.50 and $7.50, with context caching doubling the same day.

Models: [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash)

Source: [Google, Gemini Developer API pricing](https://ai.google.dev/gemini-api/docs/pricing), read 2026-10-06. Permalink: https://infro.io/ai-updates#gemini-flash-price-rise

### 6 January 2027 — OpenAI: tts-1 and tts-1-hd shut down

Shutdown. OpenAI recommends gpt-realtime-2.1-mini.

Models: [TTS-1](https://infro.io/models/openai/tts-1), [TTS-1 HD](https://infro.io/models/openai/tts-1-hd)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#tts-1-shutdown

### 14 January 2027 — OpenAI: GPT-4.1 and GPT-4o snapshots shut down

Shutdown. gpt-4.1-2025-04-14, gpt-4.1-mini-2025-04-14, gpt-4o-2024-08-06, gpt-4o-2024-11-20 and gpt-4o-mini-2024-07-18 stop answering. OpenAI recommends GPT-6 Luna.

Models: [GPT-4.1](https://infro.io/models/openai/gpt-4.1), [GPT-4.1 mini](https://infro.io/models/openai/gpt-4.1-mini), [GPT-4o](https://infro.io/models/openai/gpt-4o), [GPT-4o mini](https://infro.io/models/openai/gpt-4o-mini), [GPT-6 Luna](https://infro.io/models/openai/gpt-6-luna)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#gpt-4o-snapshots-shutdown

### 20 January 2027 — OpenAI: Legacy realtime and audio models shut down

Shutdown. gpt-realtime, gpt-realtime-mini, gpt-audio and the GPT-4o audio models stop answering. OpenAI recommends gpt-realtime-2.1 and gpt-audio-1.5.

Models: `gpt-realtime`, `gpt-realtime-mini`, `gpt-audio`, `gpt-audio-mini`, `gpt-4o-realtime`, `gpt-4o-audio`

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-19. Permalink: https://infro.io/ai-updates#openai-realtime-audio-shutdown

### 28 January 2027 — Google Cloud: Gemini 3.7 Flash retires on Vertex AI

Shutdown. Vertex AI names Gemini 3.8 Flash as the replacement.

Models: [Gemini 3.7 Flash](https://infro.io/models/google/gemini-3.7-flash), [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash)

Source: [Google Cloud, Model versions and lifecycle](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/learn/model-versions), read 2026-10-19. Permalink: https://infro.io/ai-updates#vertex-gemini-3-7-flash-retired

### 9 February 2027 — Microsoft Azure: GPT-5 2025-08-07 retires on Azure OpenAI

Shutdown. gpt-5, gpt-5-mini and gpt-5-nano version 2025-08-07 retire on Azure; OpenAI's own date for the same snapshots is 11 December 2026.

Models: `gpt-5`, `gpt-5-mini`, `gpt-5-nano`

Source: [Microsoft, Model retirement schedule](https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule), read 2026-10-19. Permalink: https://infro.io/ai-updates#azure-gpt-5-retired

### 26 February 2027 — OpenAI: whisper-1 and GPT-4o transcription shut down

Shutdown. whisper-1, the GPT-4o transcribe models and /v1/audio/translations stop answering. OpenAI recommends gpt-transcribe or gpt-live-transcribe, and realtime translations.

Models: [Whisper](https://infro.io/models/openai/whisper-1), [GPT-4o Transcribe](https://infro.io/models/openai/gpt-4o-transcribe), [GPT-4o mini Transcribe](https://infro.io/models/openai/gpt-4o-mini-transcribe), [GPT-Transcribe](https://infro.io/models/openai/gpt-transcribe)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#whisper-1-shutdown

### 15 March 2027 (earliest) — Google: Original Nano Banana leaves the Gemini API

Shutdown. gemini-2.5-flash-image stops answering on the Gemini API; Google recommends Nano Banana 2 Lite.

Models: [Nano Banana](https://infro.io/models/google/nano-banana), [Nano Banana 2 Lite](https://infro.io/models/google/nano-banana-2-lite)

Source: [Google, Gemini deprecations](https://ai.google.dev/gemini-api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#nano-banana-shutdown

### 1 April 2027 — OpenAI: GPT-5.4 nano, GPT-5.1 and GPT-5.3-Codex shut down

Shutdown. The models deprecated on 1 October 2026 stop answering.

Models: [GPT-5.4 nano](https://infro.io/models/openai/gpt-5.4-nano), [GPT-5.3 Codex](https://infro.io/models/openai/gpt-5.3-codex), [GPT-6 Luna](https://infro.io/models/openai/gpt-6-luna), [GPT-6 Sol](https://infro.io/models/openai/gpt-6-sol)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#gpt-5-1-shutdown

### 14 April 2027 — Microsoft Azure: GPT-4o and GPT-4.1 retire on Azure OpenAI

Shutdown. The 2025-04-14 GPT-4.1 family and the 2024 GPT-4o and GPT-4o mini versions retire on Azure; OpenAI's API date is 14 January 2027.

Models: [GPT-4.1](https://infro.io/models/openai/gpt-4.1), [GPT-4.1 mini](https://infro.io/models/openai/gpt-4.1-mini), [GPT-4o](https://infro.io/models/openai/gpt-4o), [GPT-4o mini](https://infro.io/models/openai/gpt-4o-mini)

Source: [Microsoft, Model retirement schedule](https://learn.microsoft.com/en-us/azure/foundry/openai/concepts/model-retirement-schedule), read 2026-10-19. Permalink: https://infro.io/ai-updates#azure-gpt-4o-retired

### 7 May 2027 (earliest) — Google: Gemini 3.1 Flash-Lite leaves the Gemini API

Shutdown. The earliest shutdown date for gemini-3.1-flash-lite; Google recommends gemini-3.5-flash-lite.

Models: [Gemini 3.1 Flash-Lite](https://infro.io/models/google/gemini-3.1-flash-lite)

Source: [Google, Gemini deprecations](https://ai.google.dev/gemini-api/docs/deprecations), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-3-1-flash-lite-shutdown

### 2 December 2027 — European Union: EU AI Act rules for high-risk systems apply

Regulation. The Commission states the rules for high-risk systems — areas such as employment, education and critical infrastructure — apply from this date.


Source: [European Commission, AI Act](https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai), read 2026-10-06. Permalink: https://infro.io/ai-updates#eu-ai-act-high-risk

## Past (45), newest first

### 17 October 2026 — DeepSeek: DeepSeek V4.1-Flash-1017 and a search API

Release. deepseek-flash moves to the 1017 version with usage and pricing unchanged. Search and Fetch APIs open at $0.75 per 1,000 calls per endpoint.

Models: [DeepSeek Flash (V4.1)](https://infro.io/models/deepseek/deepseek-flash)

Source: [DeepSeek, DeepSeek-V4.1-Flash-1017](https://api-docs.deepseek.com/news/news261017/), read 2026-10-19. Permalink: https://infro.io/ai-updates#deepseek-flash-1017

### 16 October 2026 — OpenAI: Prompt-to-voice for GPT-Live

Release. Custom voices from a text description, plus nine new voices. Creating a voice is free; speech bills as GPT-Live sessions.

Models: `gpt-live-1`

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gpt-live-prompt-to-voice

### 15 October 2026 — Anthropic: Fast mode for Opus 5.5 leaves research preview

Platform. Fast mode no longer needs an access request or the beta header, and is also available on Microsoft Foundry.

Models: [Claude Opus 5.5](https://infro.io/models/anthropic/claude-opus-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-19. Permalink: https://infro.io/ai-updates#opus-5-5-fast-mode

### 15 October 2026 (earliest) — Anthropic: Claude Haiku 4.5 retirement window opens

Deprecation. Anthropic lists Haiku 4.5 for retirement no sooner than this date.

Models: [Claude Haiku 4.5](https://infro.io/models/anthropic/claude-haiku-4.5)

Source: [Anthropic, Model deprecations](https://platform.claude.com/docs/en/about-claude/model-deprecations), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-haiku-4-5-retirement

### 14 October 2026 — OpenAI: GPT-4.1, GPT-4.1 mini, GPT-4o and GPT-4o mini deprecated

Deprecation. The dated snapshots behind these aliases shut down on 14 January 2027 (gpt-4o-2024-05-13 already on 23 October). OpenAI recommends GPT-6 Luna.

Models: [GPT-4.1](https://infro.io/models/openai/gpt-4.1), [GPT-4.1 mini](https://infro.io/models/openai/gpt-4.1-mini), [GPT-4o](https://infro.io/models/openai/gpt-4o), [GPT-4o mini](https://infro.io/models/openai/gpt-4o-mini)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#openai-gpt-4o-deprecated

### 14 October 2026 — xAI: Grok Voice 3.0

Release. Speech-to-speech at $0.08 per minute, the same as 2.0. grok-voice-latest now points to it; grok-voice-2.0 is deprecated with no date set.

Models: `grok-voice-3.0`, `grok-voice-2.0`

Source: [xAI, Introducing Grok Voice 3.0](https://x.ai/news/grok-voice-3), read 2026-10-19. Permalink: https://infro.io/ai-updates#grok-voice-3

### 14 October 2026 — Anthropic: OpenTelemetry export, in beta

Platform. Messages API requests and Managed Agents sessions can be traced with OpenTelemetry; the SDKs emit spans in seven languages.


Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-19. Permalink: https://infro.io/ai-updates#anthropic-otel

### 13 October 2026 — Google: Gemini 4 Argon generally available

Release. Paid tier only, at $2 input and $10 output through 30 November 2026; the rate doubles on 1 December.

Models: `gemini-4-argon`

Source: [Google, Release notes](https://ai.google.dev/gemini-api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-4-argon

### 10 October 2026 — Alibaba Cloud: Model Studio retires legacy Qwen and third-party models

Shutdown. The postponed retirement of older Qwen, Qwen3, audio, DeepSeek, GLM and Kimi models takes effect at 00:00 UTC+8. No replacements are named.

Models: `qwen-turbo`, `qwen-vl-max`, `qwq-plus`, `qwen3-max`, `qwen3-coder-plus`, `deepseek-v3.2`, `deepseek-r1`, `glm-4.7`, `kimi-k2-thinking`

Source: [Alibaba Cloud, Notice on the Postponed Retirement of Selected Legacy Models](https://www.alibabacloud.com/en/notice/model_studio_notice_on_the_postponed_retirement_of_selected_legacy_models_7d0), read 2026-10-19. Permalink: https://infro.io/ai-updates#alibaba-legacy-retirement

### 8 October 2026 — OpenAI: Ultrafast tier for GPT-6.1 Sol

Price. service_tier ultrafast on the Responses API: $12 input, $0.60 cached and $60 output up to 272K input tokens.

Models: [GPT-6.1 Sol](https://infro.io/models/openai/gpt-6.1-sol)

Source: [OpenAI, Pricing](https://developers.openai.com/api/docs/pricing), read 2026-10-19. Permalink: https://infro.io/ai-updates#gpt-6-1-sol-ultrafast

### 8 October 2026 — Google: Gemini 3.7 and 3.5 Flash deprecated and auto-routed

Deprecation. Requests to gemini-3.7-flash now go to gemini-3.8-flash, and gemini-3.5-flash to gemini-3.6-flash — a different model with no code change.

Models: [Gemini 3.7 Flash](https://infro.io/models/google/gemini-3.7-flash), [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash), [Gemini 3.6 Flash](https://infro.io/models/google/gemini-3.6-flash), `gemini-3.5-flash`

Source: [Google, Release notes](https://ai.google.dev/gemini-api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-3-7-flash-deprecated

### 7 October 2026 — Anthropic: Sonnet 5.5 cache reads cut to $0.10

Price. Prompt cache reads on Claude Sonnet 5.5 drop from $0.20 to $0.10 per 1M tokens. Other rates are unchanged.

Models: [Claude Sonnet 5.5](https://infro.io/models/anthropic/claude-sonnet-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-19. Permalink: https://infro.io/ai-updates#sonnet-5-5-cache-reads

### 7 October 2026 — Anthropic: Claude Haiku 5.5

Release. 1M context, 128K output. $0.10 input and $0.50 output up to 100K prompt tokens, $0.50 and $2.50 above. Rejects temperature, top_p, top_k and prefill.

Models: `claude-haiku-5-5`

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-19. Permalink: https://infro.io/ai-updates#claude-haiku-5-5

### 6 October 2026 — OpenAI: Usage tiers cut from five to three

Platform. API usage tiers are now Build, Launch and Grow.


Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#openai-usage-tiers

### 6 October 2026 — OpenAI: Decisions API in beta

Release. OpenAI's Decisions API runs on GPT-6 Luna. The pricing page lists no Decisions price yet.

Models: [GPT-6 Luna (Decisions)](https://infro.io/models/openai/gpt-6-luna-decisions)

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#openai-decisions-api-beta

### 6 October 2026 — Google: Nano Banana 2.1 generally available

Release. $1.50 input, $7.50 text output and $30 image output per 1M tokens. gemini-3.1-flash-image is deprecated, with no date set.

Models: `gemini-nano-banana-2.1`, `gemini-3.1-flash-image`

Source: [Google, Release notes](https://ai.google.dev/gemini-api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#nano-banana-2-1

### 6 October 2026 — Mistral AI: Mistral Large 4 in public preview

Release. A multimodal mixture-of-experts model with a 1M context. Mistral says open weights are coming and launch pricing is 50% off for two weeks.

Models: `mistral-large-4`

Source: [Mistral AI, Changelog](https://docs.mistral.ai/getting-started/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#mistral-large-4

### 5 October 2026 — AWS: GLM 5.3 on Amazon Bedrock

Release. Z.ai's open-weight GLM 5.3 for agentic coding is generally available on Amazon Bedrock.

Models: [GLM-5.3](https://infro.io/models/zai/glm-5.3)

Source: [AWS, GLM 5.3 by Z.ai is now generally available on Amazon Bedrock](https://aws.amazon.com/about-aws/whats-new/2026/10/amazon-bedrock-glm-5-3/), read 2026-10-19. Permalink: https://infro.io/ai-updates#glm-5-3-bedrock

### 2 October 2026 — xAI: grok-voice-transcribe-1.0 reaches end of life

Shutdown. Requests to grok-voice-transcribe-1.0 are routed to grok-voice-transcribe-2.0 at the same price.

Models: `grok-voice-transcribe-1.0`, `grok-voice-transcribe-2.0`

Source: [xAI, Release notes](https://docs.x.ai/developers/release-notes), read 2026-10-19. Permalink: https://infro.io/ai-updates#grok-voice-transcribe-1-eol

### 1 October 2026 — OpenAI: tts-1, tts-1-hd and gpt-4o-mini-tts deprecated

Deprecation. Shutdown on 6 January 2027. OpenAI recommends gpt-realtime-2.1-mini.

Models: [TTS-1](https://infro.io/models/openai/tts-1), [TTS-1 HD](https://infro.io/models/openai/tts-1-hd), `gpt-4o-mini-tts-2025-03-20`, `gpt-4o-mini-tts-2025-12-15`

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#openai-tts-1-deprecated

### 1 October 2026 — OpenAI: GPT-5.4 nano, GPT-5.1 and GPT-5.3-Codex deprecated

Deprecation. Shutdown on 1 April 2027. OpenAI recommends GPT-6 Luna for GPT-5.4 nano and GPT-6 Sol for GPT-5.3-Codex.

Models: [GPT-5.4 nano](https://infro.io/models/openai/gpt-5.4-nano), [GPT-5.3 Codex](https://infro.io/models/openai/gpt-5.3-codex)

Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#openai-gpt-5-1-deprecated

### 30 September 2026 — Mistral AI: OCR 4.0 retired after one day's notice

Shutdown. Announced on 29 September for the next day. Mistral recommends OCR 4.1 at the same price.


Source: [Mistral AI, Changelog](https://docs.mistral.ai/getting-started/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#mistral-ocr-4-retired

### 30 September 2026 — Anthropic: Claude Sonnet 4.5 retirement announced

Deprecation. Sonnet 4.5 retires on 30 November 2026; Anthropic names Sonnet 5.5 as the replacement.

Models: [Claude Sonnet 5.5](https://infro.io/models/anthropic/claude-sonnet-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-sonnet-4-5-deprecated

### 29 September 2026 — OpenAI: Computer use in the Agents API

Platform. The Agents API gains computer use.


Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#openai-agents-computer-use

### 29 September 2026 — OpenAI: GPT-6.1 Sol, and an ultrafast tier for Astra

Release. GPT-6.1 Sol at $2 input and $10 output with cached input at $0.10. GPT-6 Astra gains an ultrafast service tier on the Responses API.

Models: [GPT-6.1 Sol](https://infro.io/models/openai/gpt-6.1-sol), [GPT-6 Astra](https://infro.io/models/openai/gpt-6-astra)

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#gpt-6-1-sol

### 28 September 2026 — Anthropic: Claude Sonnet 5.5

Release. $2 input and $10 output, which the launch note calls introductory pricing. Turning off up-front thinking is now thinking type between_tools, not disabled.

Models: [Claude Sonnet 5.5](https://infro.io/models/anthropic/claude-sonnet-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-sonnet-5-5

### 27 September 2026 — Perplexity: Sonar Chat Completions support ends

Shutdown. Synchronous and streaming requests are reformulated as Agent API calls, rolling out by model. Asynchronous Sonar requests must move to the Agent API's background mode.

Models: [Sonar](https://infro.io/models/perplexity/sonar), [Sonar Pro](https://infro.io/models/perplexity/sonar-pro)

Source: [Perplexity, Migrate from Sonar to the Agent API](https://docs.perplexity.ai/docs/agent-api/migrate-from-sonar/overview), read 2026-10-06. Permalink: https://infro.io/ai-updates#sonar-chat-completions-end

### 24 September 2026 — OpenAI: Sora 2 and the Videos API shut down

Shutdown. OpenAI ended Sora 2 and the Videos API.


Source: [OpenAI, Deprecations](https://developers.openai.com/api/docs/deprecations), read 2026-10-17. Permalink: https://infro.io/ai-updates#sora-2-shutdown

### 24 September 2026 — Anthropic: Billing resumes for early refusals

Price. Refusals that arrive before any output are billed again in three categories.


Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#anthropic-refusal-billing

### 23 September 2026 — Anthropic: Cache diagnostics generally available

Platform. Anthropic made its prompt-cache diagnostics generally available on the API.


Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#anthropic-cache-diagnostics

### 22 September 2026 — OpenAI: GPT-6 Sol and GPT-6 Luna

Release. Sol at $2 input, $0.20 cached and $10 output; Luna at $0.10, $0.01 and $0.50.

Models: [GPT-6 Sol](https://infro.io/models/openai/gpt-6-sol), [GPT-6 Luna](https://infro.io/models/openai/gpt-6-luna)

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#gpt-6-sol-luna

### 22 September 2026 — Google: Gemini 3.8 Flash TTS and Flash-Lite TTS

Release. Both text-to-speech models are generally available, with the new /v1beta/voices endpoint.

Models: `gemini-3.8-flash-tts`, `gemini-3.8-flash-lite-tts`

Source: [Google, Release notes](https://ai.google.dev/gemini-api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gemini-3-8-tts

### 22 September 2026 — Anthropic: Claude Opus 5.5

Release. $4 input and $20 output, below Opus 5's $5 and $25. Thinking cannot be disabled: both disabled and enabled thinking settings return a 400.

Models: [Claude Opus 5.5](https://infro.io/models/anthropic/claude-opus-5.5)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-opus-5-5

### 21 September 2026 — xAI: Grok 4.7

Release. $2 input, $0.50 cached and $6 output below 200K prompt tokens; $4 and $12 above. 500K-token context, reasoning effort from low to xhigh.

Models: [Grok 4.7](https://infro.io/models/xai/grok-4.7)

Source: [xAI, grok-4.7](https://docs.x.ai/docs/models/grok-4.7), read 2026-10-06. Permalink: https://infro.io/ai-updates#grok-4-7

### 15 September 2026 — TypeSafe: Jev, a System One model

Release. Answers typed questions — yes/no, one of a set, a rubric score — with probabilities instead of text. $0.042 per 1M input tokens; output is free.

Models: [Jev 1.13](https://infro.io/models/typesafe/jev-1.13)

Source: [TypeSafe, Introducing System One Models & Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev), read 2026-10-06. Permalink: https://infro.io/ai-updates#typesafe-jev

### 14 September 2026 — Anthropic: Compaction on demand, in beta

Platform. A top-level compaction parameter returns a signed compaction block that later requests send in place of the summarized messages.


Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-19. Permalink: https://infro.io/ai-updates#anthropic-compaction

### 10 September 2026 — OpenAI: Agents API in public beta

Platform. OpenAI opened the Agents API to everyone as a public beta.


Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#openai-agents-api-beta

### 10 September 2026 — OpenAI: GPT-Live 1 generally available

Release. Voice sessions at $0.05 per minute, billed per second; the backend model and tool usage bill separately.

Models: `gpt-live-1`

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gpt-live-1

### 10 September 2026 — DeepSeek: DeepSeek V4.1-Flash

Release. Served as deepseek-flash, with off-peak prices at half the peak rate. The older deepseek-v4-flash names now route to it.

Models: [DeepSeek Flash (V4.1)](https://infro.io/models/deepseek/deepseek-flash), [DeepSeek V4 Flash](https://infro.io/models/deepseek/deepseek-v4-flash)

Source: [DeepSeek, DeepSeek-V4.1-Flash: Smarter, Faster, More Efficient](https://api-docs.deepseek.com/news/news260910/), read 2026-10-06. Permalink: https://infro.io/ai-updates#deepseek-v4-1-flash

### 8 September 2026 — OpenAI: GPT Image 2.5 Sunburst and Flare

Release. Two image models with new xhigh and max quality settings, billed at GPT Image 2's token rates.

Models: [GPT Image 2.5 Sunburst](https://infro.io/models/openai/gpt-image-2.5-sunburst), [GPT Image 2.5 Flare](https://infro.io/models/openai/gpt-image-2.5-flare)

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-19. Permalink: https://infro.io/ai-updates#gpt-image-2-5

### 8 September 2026 — AWS: GPT-6 Astra on Amazon Bedrock

Release. AWS made OpenAI's GPT-6 Astra generally available on Amazon Bedrock.

Models: [GPT-6 Astra](https://infro.io/models/openai/gpt-6-astra)

Source: [AWS, OpenAI GPT-6 Astra is now generally available on Amazon Bedrock](https://aws.amazon.com/about-aws/whats-new/2026/09/openai-gpt-6-astra-on-amazon-bedrock/), read 2026-10-06. Permalink: https://infro.io/ai-updates#gpt-6-astra-bedrock

### 5 September 2026 — Alibaba Cloud: qwen3.8-max now points at the 0902 snapshot

Platform. The unpinned qwen3.8-max alias moved to qwen3.8-max-0902 with pricing unchanged. Requests naming the alias got a different model with no code change.

Models: [Qwen3.8 Max](https://infro.io/models/qwen/qwen3.8-max)

Source: [Alibaba Cloud, Update Notice for Qwen3.8-Max Models](https://www.alibabacloud.com/en/notice/model_studio_update_notice_for_qwen38max_models_863), read 2026-10-06. Permalink: https://infro.io/ai-updates#qwen3-8-max-alias

### 3 September 2026 — OpenAI: GPT-6 Astra

Release. $10 input and $50 output; above 272K input tokens the whole request bills 2x input and 1.5x output. No custom temperature, top_p or logprobs; tool calling needs the Responses API.

Models: [GPT-6 Astra](https://infro.io/models/openai/gpt-6-astra)

Source: [OpenAI, Changelog](https://developers.openai.com/api/docs/changelog), read 2026-10-06. Permalink: https://infro.io/ai-updates#gpt-6-astra

### 2 September 2026 — Google: Gemini 3.8 Flash is generally available

Release. 1M-token context, 64K output, $0.75 input and $3.75 output until 31 December 2026. Remove temperature, top_p and top_k; thinking_level replaces thinking_budget.

Models: [Gemini 3.8 Flash](https://infro.io/models/google/gemini-3.8-flash)

Source: [Google, What's new in Gemini 3.8 Flash](https://ai.google.dev/gemini-api/docs/latest-model), read 2026-10-06. Permalink: https://infro.io/ai-updates#gemini-3-8-flash

### 1 September 2026 — Anthropic: Claude Fable 5.1

Release. $10 input and $50 output, with cache reads cut to $0.25. 1M-token context and 128K output. Forced tool use returns a 400; Anthropic recommends auto with strict tools.

Models: [Claude Fable 5.1](https://infro.io/models/anthropic/claude-fable-5.1)

Source: [Anthropic, API release notes](https://platform.claude.com/docs/en/release-notes/api), read 2026-10-06. Permalink: https://infro.io/ai-updates#claude-fable-5-1