LLM Availability Overview

5 min read

LLM Availability Overview

Unique can connect Spaces to Azure OpenAI and third-party LLMs (via LiteLLM). Use the matrices below to pick a LanguageModelName, check Chat/Translation quality support, retirement, and (for Azure) data-residency SKUs.

Setup: In a Space, choose an AI model from the dropdown, or under Advanced Settings → Language Model paste a value from the first column of the matrices. Extra configuration notes: Language model setup. Confirm activated models with support or your contract.

Per-client availability: The tables below list models Unique can offer across the platform. What is actually deployed for a given client can differ — for example if no Google (Gemini / Vertex) credentials have been configured for that tenant. Each model provider must be contractually agreed with the client. Some models are only deployed on request (for example because they can incur very high costs). Clients can always reach out to Customer Success for a full extract of the models deployed in their environment.

Retirement dates are taken from the vendor pages listed under Vendor retirement schedules (re-checked 2026-07-30).

Model Availability Matrix

Legend (Unique AI Chat)

Symbol	Meaning
⚪	Not yet validated
🔴	Not suited
🟡	Functionally working — no quality support
🟢	Passed extensive quality testing

Azure Deployment

Every model in the Azure matrix below is available via Global Standard (inference across Microsoft regions, including US; default for US tenants). The Residency column lists additional SKUs when available:

CH — Standard (Switzerland)
EU — Data Zone Standard (Europe)
— — Global Standard only

LanguageModelName	Model	Retirement · Lifecycle	Unique release	Chat	Residency
`AZURE_GPT_41_2025_0414`	`gpt-4.1` / 2025-04-14	2027-04-14 · Deprecated	2025.24	⚪	CH · EU
`AZURE_GPT_41_MINI_2025_0414`	`gpt-4.1-mini` / 2025-04-14	2027-04-14 · Deprecated	2025.24	⚪	CH · EU
`AZURE_GPT_41_NANO_2025_0414`	`gpt-4.1-nano` / 2025-04-14	2026-10-14 · Deprecated	2025.24	⚪	EU
`AZURE_GPT_4o_2024_0513`	`gpt-4o` / 2024-05-13	2026-10-01 · Deprecated	2025.12	🟡	EU
`AZURE_GPT_4o_2024_0806`	`gpt-4o` / 2024-08-06	2027-04-14 · Deprecated	2025.12	🟡	EU
`AZURE_GPT_4o_2024_1120`	`gpt-4o` / 2024-11-20	2027-04-14 · Deprecated	2025.18	🟢	CH · EU
`AZURE_GPT_4o_MINI_2024_0718`	`gpt-4o-mini` / 2024-07-18	2027-04-14 · Deprecated	before 2025.12	🔴	EU
`AZURE_GPT_5_2025_0807`	`gpt-5` / 2025-08-07	2027-02-09 · GA	2025.34	🟢	EU
`AZURE_GPT_5_MINI_2025_0807`	`gpt-5-mini` / 2025-08-07	2027-02-09 · GA	2025.34	🟡	EU
`AZURE_GPT_5_NANO_2025_0807`	`gpt-5-nano` / 2025-08-07	2027-02-09 · GA	2025.34	🟡	EU
`AZURE_GPT_51_2025_1113`	`gpt-5.1` / 2025-11-13	2027-05-15 · GA	2025.44	🟢	EU
`AZURE_GPT_52_2025_1211`	`gpt-5.2` / 2025-12-11	2027-06-08 · GA	2025.52	🟢	—
`AZURE_GPT_54_2026_0305`	`gpt-5.4` / 2026-03-05	2027-09-02 · GA	2026.12	🟢	EU
`AZURE_GPT_54_PRO_2026_0305`	`gpt-5.4-pro` / 2026-03-05	2027-09-07 · GA	2026.12	🔴	—
`AZURE_GPT_55_2026_0424`	`gpt-5.5` / 2026-04-24	2027-10-26 · GA	2026.18	🟢	EU
`AZURE_GPT_56_LUNA_2026_0709`	`gpt-5.6-luna` / 2026-07-09	2028-01-11 · GA	2026.30	🟢	EU
`AZURE_GPT_56_SOL_2026_0709`	`gpt-5.6-sol` / 2026-07-09	2028-01-11 · GA	2026.30	🟢	EU
`AZURE_GPT_56_TERRA_2026_0709`	`gpt-5.6-terra` / 2026-07-09	2028-01-11 · GA	2026.30	🟢	EU
`AZURE_o1_2024_1217`	`o1` / 2024-12-17	2026-10-21 · Deprecated	2025.14	🟡	EU
`AZURE_o3_2025_0416`	`o3` / 2025-04-16	2026-10-21 · GA	2025.20	🟡	EU
`AZURE_o3_MINI_2025_0131`	`o3-mini` / 2025-01-31	2026-10-01 · Deprecated	2025.14	🔴	EU
`AZURE_o4_MINI_2025_0416`	`o4-mini` / 2025-04-16	2026-10-16 · Deprecated	2025.20	🟡	EU

LiteLLM Deployment

Region / data residency by provider:

Anthropic, Together AI — globally available only (no CH / EU data-zone restriction).
Vertex AI (Claude) — EU only possible

Anthropic

Globally available only.

LanguageModelName	Model	Retirement	Unique release	Chat
`litellm:anthropic-claude-haiku-4-5`	`claude-haiku-4-5`	Not sooner than 2026-10-15	2025.44	⚪
`litellm:anthropic-claude-opus-4-1`	`claude-opus-4-1`	Deprecated — retires 2026-08-05	2025.40	🟡
`litellm:anthropic-claude-opus-4-5`	`claude-opus-4-5`	Not sooner than 2026-11-24	2025.50	🟡
`litellm:anthropic-claude-opus-4-6`	`claude-opus-4-6`	Not sooner than 2027-02-05	2026.08	🟢
`litellm:anthropic-claude-opus-4-7`	`claude-opus-4-7`	Not sooner than 2027-04-16	2026.18	🟢
`litellm:anthropic-claude-opus-4-8`	`claude-opus-4-8`	Not sooner than 2027-05-28	2026.24	🟢
`litellm:anthropic-claude-opus-5`	`claude-opus-5`	Not sooner than 2027-07-24	2026.32	🟡
`litellm:anthropic-claude-sonnet-4-5`	`claude-sonnet-4-5`	Not sooner than 2026-09-29	2025.40	🟡
`litellm:anthropic-claude-sonnet-4-6`	`claude-sonnet-4-6`	Not sooner than 2027-02-17	2026.10	🟡
`litellm:anthropic-claude-sonnet-5`	`claude-sonnet-5`	Not sooner than 2027-06-30	2026.28	🟢

Vertex AI (Claude)

EU only.

LanguageModelName	Model	Retirement	Unique release	Chat
`litellm:vertex-claude-opus-4-6`	`claude-opus-4-6`	Not sooner than 2027-02-05	2026.20	🟢
`litellm:vertex-claude-opus-4-7`	`claude-opus-4-7`	Not sooner than 2027-04-16	2026.20	🟢
`litellm:vertex-claude-opus-4-8`	`claude-opus-4-8`	Not sooner than 2027-05-28	2026.24	🟢
`litellm:vertex-claude-opus-5`	`claude-opus-5`	Not sooner than 2027-07-24	2026.32	🟡
`litellm:vertex-claude-sonnet-4-6`	`claude-sonnet-4-6`	Not sooner than 2027-02-17	2026.20	🟢
`litellm:vertex-claude-sonnet-5`	`claude-sonnet-5`	Not sooner than 2027-06-30	2026.28	🟢

Gemini

LanguageModelName	Model	Retirement	Unique release	Chat
`litellm:gemini-2-5-flash`	`gemini-2-5-flash`	2026-10-16	2025.20	🟡
`litellm:gemini-2-5-flash-lite`	`gemini-2-5-flash-lite`	2026-10-16	2025.46	🟡
`litellm:gemini-2-5-pro`	`gemini-2-5-pro`	2026-10-16	2025.28	🟢
`litellm:gemini-3-1-pro-preview`	`gemini-3-1-pro-preview`	n/a	2026.10	🟡
`litellm:gemini-3-6-flash`	`gemini-3-6-flash`	n/a	2026.34	⚪
`litellm:gemini-3-flash-preview`	`gemini-3-flash-preview`	n/a	2025.52	🟢

OpenAI (via LiteLLM)

For Unique AI Chat, prefer the Azure deployment of the same model when available. LiteLLM OpenAI aliases remain valid config keys. Retirement dates below mirror the Azure Foundry schedule for the matching model/version.

On request only: Pro and deep-research models (*-pro*, *-deep-research, including their thinking variants) are available only on request — confirm with support or your contract before enabling.

LanguageModelName	Model	Retirement	Unique release	Chat
`litellm:openai-gpt-4-1-mini`	`gpt-4-1-mini`	2027-04-14	2025.20	🟡
`litellm:openai-gpt-4-1-nano`	`gpt-4-1-nano`	2026-10-14	2025.20	🔴
`litellm:openai-gpt-5`	`gpt-5`	2027-02-09	2025.34	🟡
`litellm:openai-gpt-5-mini`	`gpt-5-mini`	2027-02-09	2025.34	🟡
`litellm:openai-gpt-5-nano`	`gpt-5-nano`	2027-02-09	2025.34	🟡
`litellm:openai-gpt-5-1`	`gpt-5-1`	2027-05-15	2025.44	⚪
`litellm:openai-gpt-5-1-thinking`	`gpt-5-1` (thinking)	2027-05-15	2025.44	⚪
`litellm:openai-gpt-5-2`	`gpt-5-2`	2027-06-08	2025.52	🟢
`litellm:openai-gpt-5-2-thinking`	`gpt-5-2` (thinking)	2027-06-08	2025.52	⚪
`litellm:openai-gpt-5-2-pro`	`gpt-5-2-pro`	2027-06-08	on request	⚪
`litellm:openai-gpt-5-2-pro-thinking`	`gpt-5-2-pro` (thinking)	2027-06-08	on request	⚪
`litellm:openai-gpt-5-4`	`gpt-5-4`	2027-09-02	2026.12	🟢
`litellm:openai-gpt-5-4-thinking`	`gpt-5-4` (thinking)	2027-09-02	2026.12	⚪
`litellm:openai-gpt-5-4-pro`	`gpt-5-4-pro`	2027-09-07	on request	🔴
`litellm:openai-gpt-5-4-pro-thinking`	`gpt-5-4-pro` (thinking)	2027-09-07	on request	⚪
`litellm:openai-gpt-5-5`	`gpt-5-5`	2027-10-26	2026.18	🔴
`litellm:openai-gpt-5-6-luna`	`gpt-5-6-luna`	2028-01-11	2026.30	🟢
`litellm:openai-gpt-5-6-sol`	`gpt-5-6-sol`	2028-01-11	2026.30	🟢
`litellm:openai-gpt-5-6-terra`	`gpt-5-6-terra`	2028-01-11	2026.30	🟢
`litellm:openai-o1`	`o1`	2026-10-21	2025.20	🟡
`litellm:openai-o3`	`o3`	2026-10-21	2025.20	🟡
`litellm:openai-o3-pro`	`o3-pro`	2026-12-17	on request	🔴
`litellm:openai-o3-deep-research`	`o3-deep-research`	2026-12-26	on request	⚪
`litellm:openai-o4-mini`	`o4-mini`	2026-10-16	2025.20	🟡
`litellm:openai-o4-mini-deep-research`	`o4-mini-deep-research`	n/a	on request	⚪

Together AI

Globally available only. Retirement: no fixed date; deprecation is usually communicated 2–3 weeks in advance (Together deprecations).

LanguageModelName	Model	Unique release	Chat
`litellm:deepseek-v4-pro`	`deepseek-v4-pro`	2026.28	🟢
`litellm:glm-5.2`	`glm-5.2`		🟢
`litellm:gpt-oss-120b`	`gpt-oss-120b`	On request only	⚪
`litellm:llama-3-3-70b-instruct-turbo`	`llama-3-3-70b-instruct-turbo`	2025.20	🔴
`litellm:kimi-k3`	`kimi-k3`	2025.32	⚪

Retired Models

LanguageModelName	Model	Retirement	Replacement
`AZURE_GPT_4_0613`	`gpt-4` / 0613	2025-06-06	`AZURE_GPT_4o_2024_1120`
`AZURE_GPT_4_32K_0613`	`gpt-4-32k` / 0613	2025-06-06	`AZURE_GPT_4o_2024_1120`
`AZURE_GPT_4_TURBO_2024_0409`	`gpt-4-turbo` / 2024-04-09	2025-06-06	`AZURE_GPT_4o_2024_1120`
`litellm:gemini-2-5-flash-preview-04-17`	`gemini-2-5-flash-preview-04-17`	2025-07	`litellm:gemini-2-5-flash`
`litellm:gemini-2.5-pro-preview-06-05`	`gemini-2.5-pro-preview-06-05`	2025-12-02	`litellm:gemini-2-5-pro`
`AZURE_GPT_35_TURBO_0125`	`gpt-35-turbo` / 0125	2025-11-11	`AZURE_GPT_4o_2024_1120`
`AZURE_o1_MINI_2024_0912`	`o1-mini` / 2024-09-12	2025-11-17	`AZURE_o4_MINI_2025_0416`
`litellm:anthropic-claude-3-7-sonnet`	`claude-3-7-sonnet`	2026-02-19	`litellm:anthropic-claude-sonnet-4-5`
`litellm:anthropic-claude-3-7-sonnet-thinking`	`claude-3-7-sonnet-thinking`	2026-02-19	`litellm:anthropic-claude-sonnet-4-5`
`litellm:gemini-3-pro-preview`	`gemini-3-pro-preview`	2026-03-09	`litellm:gemini-3-1-pro-preview`
`litellm:deepseek-r1`	`deepseek-r1`	2026-05-14	—
`litellm:deepseek-v3-1`	`deepseek-v3-1`	2026-05-14	—
`litellm:anthropic-claude-opus-4`	`claude-opus-4`	2026-06-15	`litellm:anthropic-claude-opus-4-6`
`litellm:anthropic-claude-sonnet-4`	`claude-sonnet-4`	2026-06-15	`litellm:anthropic-claude-sonnet-4-6`
`litellm:qwen-3-235B-A22B`	`qwen-3-235B-A22B`	2026-07-10	—
`litellm:qwen-3-235B-A22B-thinking`	`qwen-3-235B-A22B-thinking`	2026-07-10	—
`litellm:glm-5.1`	`glm-5.1`	2026-07-10	`litellm:glm-5.2`
`litellm:grok-4-1-fast-non-reasoning`	`grok-4-1-fast-non-reasoning`	2026-05-15	—
`litellm:grok-4-1-fast-reasoning`	`grok-4-1-fast-reasoning`	2026-05-15	—

Vendor retirement schedules

Azure: Model retirement schedule
Google Gemini: Deprecations
Anthropic: Model deprecations
Together AI: Deprecations

Recent changes

Older entries: LLM Availability — Change Log archive.

Date	Change Summary
2026-08-04	New model: `litellm:gemini-3-6-flash` (Google AI Studio track only, no Vertex entry)
2026-07-30	Clarified per-client availability: tables are platform catalog; actual deployment varies by contract/config; CS for full Auszug
2026-07-30	Documented LiteLLM: `openai-gpt-5-2-thinking`
2026-07-30	Documented LiteLLM: `openai-gpt-5-1` (+ thinking), `openai-gpt-5-2-pro` (+ thinking), `openai-gpt-5-4-thinking`, `openai-gpt-5-4-pro-thinking`, `openai-o3-deep-research`, `openai-o4-mini-deep-research`, `gpt-oss-120b`. Pro + deep-research marked on request.
2026-07-30	New model: `litellm:kimi-k3`
2026-07-26	New models: `litellm:anthropic-claude-opus-5` `litellm:vertex-claude-opus-5`
2026-07-15	Model retirement: `litellm:qwen-3-235B-A22B` `litellm:qwen-3-235B-A22B-thinking` `litellm:glm-5.1` `litellm:deepseek-r1` `litellm:deepseek-v3-1` `litellm:grok-4-1-fast-non-reasoning` `litellm:grok-4-1-fast-reasoning`
2026-07-14	New models: `litellm:openai-gpt-5-6-luna` `litellm:openai-gpt-5-6-sol` `litellm:openai-gpt-5-6-terra`
2026-07-07	New models: `litellm:anthropic-claude-sonnet-5` `litellm:vertex-claude-sonnet-5` `litellm:deepseek-v4-pro`
2026-05-26	Retirement of `litellm:anthropic-claude-opus-4` `litellm:anthropic-claude-sonnet-4`
2026-05-26	Claude models provided via Vertex AI added
2026-05-12	DataZoneStandard via Azure: `AZURE_GPT_54_2026_0305` `AZURE_GPT_55_2026_0424`