Azure OpenAI in Foundry Models vs DeepInfra

Azure OpenAI in Foundry Models 95

Model reading. Microsoft Entra ID tokens with Azure roles (the Cognitive Services OpenAI User role for calls), managed identities, and key authentication that can be disabled per resource, or two regenerable resource keys in the api-key header. No documented route puts the key in a URL query string (30). Prompts, completions, embeddings and training data aren't used to train foundation models without the customer's permission or instruction, per the data privacy page updated 4 August 2026 (20). Flagged prompts and completions may be stored in the customer's geography for human abuse review, with no retention period stated, and approved customers can turn that storage off through modified abuse monitoring. Responses API data is kept 30 days by default unless store is false (12 of 15). Azure Monitor resource logs and the subscription activity log, as read in the Foundry fine-tuning dossier (15). MSRC coordinated disclosure policy and Microsoft's bounty programme, both named in microsoft.com's security.txt, and Azure OpenAI is listed in FedRAMP High and DoD IL2 scope in Azure public regions. The security.txt passed its Expires date on 2026-09-23 and was still expired on 6 November, which the Bedrock and S3 dossiers mark down by 2, applied here too (18 of 20, down from 20 on 6 October for consistency).

DeepInfra 64

Model reading. Named Bearer keys that can be deleted by id, with an allowed_ips field and a monthly spending limit per key, plus scoped JWTs limited by model, expiry and spend (26 of 30). Less 10 because GET /v1/scoped-jwt?jwtoken= is still the documented way to inspect a JWT and puts a live token in the query string. The two deprecated routes that carried the API token in the URL path are no longer in the OpenAPI document (16 of 30). The terms say Customer Data is not used to train, fine-tune or improve any model. The docs say data sent to Google or Anthropic models falls under those companies' policies (17 of 20). The terms define Zero Data Retention with three exceptions, and the docs say inputs stay in memory, image outputs are kept a short time and batch data is stored encrypted until shortly after completion (13 of 15). Usage per key, request costs and deployment logs are in the API. No account audit log was found (8 of 15). The trust centre at trust.deepinfra.com now loads and lists SOC 2, ISO 27001 and HIPAA as compliant, with reports and policies on request, and the privacy policy says the same. No security.txt, disclosure policy or bug bounty was found (10 of 20, as other listings with certification claims and no disclosure route).

Azure OpenAI in Foundry Models 86

Model reading. GPT-6.1 Terra released on Azure on 26 October 2026 and a gpt-chat-latest preview of 7 October was added, after GPT-6.1 Sol on 29 September (30). GA models get a retirement date 18 months after launch, set at release and readable through the Models API, with at least 60 days' email and Service Health notice, and previews at least 30 days, so a GA retirement is known more than six months ahead (12 of 12). Many shutdown dates in the 90 days to 6 November, most of them previews or partner models, among them DeepSeek-R1 (13 August), MAI-Image-2 and 2e (15 August), model-router previews (30 August), MAI-Transcribe-1 (15 September), MAI-Image-2.5 previews (1 October), the 2025-03-20 audio models and sora-2 (15 October), a Cohere Command A Plus preview (16 October) and gpt-image-1 (23 October), each in the schedule with a replacement where one exists (4 of 8). The Foundry what's new page, still on August when read on 6 November, plus Microsoft Q&A and paid support; no Azure OpenAI what's new was found (10 of 15). The SDKs are OpenAI's repositories, replies not sampled (5 of 10). Current official SDKs, openai 3.46.0 on PyPI (5 November) and 7.53.0 on npm, plus Microsoft's @azure/openai 2.0.0 companion (15). Supported runtimes stated, Python 3.9 or later and Node.js 20 or later, though openai 3.46.0 requires Python 3.10 (10).

DeepInfra 61

Model reading. The docs repository changed on 26 October 2026, and on 22 October documented Nemotron 3.5 image embeddings and reranker scores (30). Deprecated models get at least one week's notice by email, then requests are forwarded to a replacement (5 of 12). Churn. In the 90 days to 2 November four redirect dates passed (8, 9, 12 and 25 October, five models), Kimi-K2.7-Code was deprecated before 28 September, and three more are scheduled for 2 and 3 November. Three or more removal dates in 90 days give 0, as groq and fireworks-inference were read (0 of 8, was 4 when the five were only scheduled). No changelog. Support is by Discord and a feedback address, and the docs repository is public (7 of 15). Replies on the SDK repositories were not examined (3 of 10). Python deepinfra 0.3.0 from 12 August 2026 with a changelog. The Node repository tagged 2.1.0 on 1 September 2026, and npm couldn't be read on 2 November, so its 8 October reading of 2.0.2 from 8 May 2024 stands (9 of 15). CI workflows in both SDK repositories (7 of 10). 30 + 5 + 0 + 7 + 3 + 9 + 7 = 61.

Azure OpenAI in Foundry Models 98

Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). Tool use with parallel calls and a forced call by naming the function in tool_choice, 18 of 20 because gpt-6.1-sol calls tools only through the Responses API (18). Strict structured outputs (15). Prompt caching from 1,024 tokens, cached input at 0.1x on GPT-6 ($1 per million on Astra) and cache writes at 1.25x from GPT-5.6 on, with explicit breakpoints (15). 1,050,000-token context on GPT-6 and GPT-6.1, input capped at 922,000 (15). Batch at 50% off Global Standard with a 24-hour target (10). The docs use OpenAI's own SDKs for Python, JavaScript, C#, Go and Java against the v1 endpoint (10). 429 and quota errors documented with retry headers (15). Calls name a deployment rather than a model, so each model must be deployed in the resource first, which the model checklist doesn't score.

DeepInfra 77

Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). Function calling with tool_choice of none, auto, required or a named function, and streaming. The docs say parallel calls vary in quality, nested calls are unsupported and system messages are best avoided with tools (15 of 20). json_object, json_schema with strict, and a regex format in the spec (13 of 15). Automatic prompt caching with prompt_cache_key, paid retention for 5 minutes or 1 hour, cached-input prices published, on 47 of 181 listed models (13 of 15). Context of 1,048,576 tokens on DeepSeek V4, Kimi K3 and GLM 5.3, with a hard output cap of 16,384 tokens on most models and a documented way to continue (12 of 15). Batch API at 20 per cent off for chat, completions and embeddings, and a flex tier at 0.8x (10). The OpenAI and Anthropic clients work with a changed base URL. The Python package deepinfra 0.3.0 dates from 12 August 2026, and npm listed 2.0.2 from May 2024 on 8 October (7 of 10). 429 carries an engine_overloaded code. Native routes return a bare error string, and no list of codes was found (7 of 15).

Azure OpenAI in Foundry Models 87

OpenAPI 3.2 spec for the v1 data plane in Azure/azure-rest-api-specs (azure-v1-v1-generated.json, 71 paths and 852 schemas), linked from the v1 API page (25). No llms.txt at learn.microsoft.com (404). Learn pages offer Copy Markdown and the Markdown sources are public in MicrosoftDocs/azure-ai-docs, last merged 5 October (5 of 10). The models page lists each model's capabilities, context and API support, but the model choice guide, dated 19 May 2026, doesn't cover GPT-6 (15 of 20). Structured outputs with strict: true on schemas and function tools, though not together with parallel function calls (15). Examples in Python, C#, JavaScript, Go, Java and REST, and an error table on the quotas page with a fix for 429 and other codes (15). The v1 path replaces monthly api-version values, with preview capabilities opted into by header, a dated API version changelog and dated model versions. On 6 November the Foundry what's new still covered August, though updated on 5 November, and no Azure OpenAI what's new was found at learn.microsoft.com/en-us/azure/foundry/openai/whats-new, which answered 404 (12 of 15).

DeepInfra 69

Model reading. Public OpenAPI 3.1.0 document at api.deepinfra.com/openapi.json with 131 paths, 157 operations and 257 schemas on 2 November 2026. llms.txt now links that document, though the docs host's own copy still returns Mintlify's sample plant store document, and the spec lists chat at /v1/chat/completions while every docs example uses /v1/openai/chat/completions (22 of 25). llms.txt with 224 links and a Markdown copy of every docs page (10). Guide pages for tool calling, structured output, caching, batch and service tiers say when to use each. Many reference operations have a name and no description (14 of 20). Typed parameters with 49 enums, 31 chat properties and two required (11 of 15). Curl, Python and JavaScript examples on each page. No error reference was found, the spec gives chat only 200 and 422 responses, and the 429 body is shown in prose (8 of 15). No changelog was found. The spec version reads 1.0.0, and the docs repository on GitHub has a public commit history (4 of 15).

Azure OpenAI in Foundry Models 83

Closed service under the Microsoft Product Terms, with the v1 spec under MIT and the OpenAI SDKs under Apache-2.0 (15). The data privacy page agrees with the product terms on no training and states where data is processed and stored by deployment type, but gives no retention period for the human-review store and sits beside a 30-day default for Responses data (24 of 30). A published lifecycle policy and a retirement schedule with dates and replacements, but on 6 November the schedule listed gpt-realtime-mini 2025-10-06 and 2025-12-15 twice each with different retirement dates, and no GPT-6 model yet (18 of 20, down from 20). Data locations stated for Standard, Data Zone and Global deployments, and a public sub-processor list per the 5 October Azure Speech check, not re-read (20).

DeepInfra 68

Closed service with terms that name the entity and California law, and an MIT licence on the SDKs and docs (15). The terms (last modified 15 October 2026), the privacy policy and the data privacy page agree on no storage and no training, the terms say their retention clause controls, and personal information is removed 30 days after account deletion. The terms now point to a DPA on the trust centre, which lists it among documents available on request, and the entity is written Deep Infra Inc. in the terms and DeepInfra, Inc. in the privacy policy (21 of 30). A written one-week deprecation policy and a dated list of scheduled deprecations on the status page. No record of past removals was found (14 of 20). The sub-processor list names Stripe, Amazon Web Services and Google Cloud Platform, as does the trust centre. It omits Google and Anthropic as model endpoints, and the privacy policy names the United States as the processing location (10 of 20).

Azure OpenAI in Foundry Models 80

Azure status page with post-incident reviews, listing Azure OpenAI Service, Foundry Models and Foundry Agent Service as components (20). One incident in the 90 days to 6 November 2026 names Azure OpenAI, and no review after it does, intermittent request failures, higher latency and HTTP 5xx errors in Sweden Central from 10:03 to 15:58 UTC on 29 September 2026 (tracking 7QL5-Z50), counted as one major outage as in the Foundry fine-tuning dossier. The 23 July West US review and the 30 September gateway incident don't name it (10). Quota tiers from Free to Tier 6 with TPM and RPM per model and deployment type, 1,000 RPM and 1,000,000 TPM for each GPT-6 model on Global Standard at Tier 1 (15). 429 responses carry retry-after-ms, every response carries x-ratelimit-* headers, and the quota guide gives exponential backoff with jitter and SDK retry settings (15). The models page says models sold by Azure are covered by Azure service-level agreements and links the Online Services SLA. The consolidated SLA document wouldn't download on 6 October and wasn't retried on 6 November, so the clause rests on Microsoft's docs, read the same way as the 5 October Azure Speech check (10). The v1 API is GA and the GPT-6 models carry no preview label (10).

DeepInfra 75

Hosted reading. Own status page at status.deepinfra.com with 90-day day-by-day history for the API, the website and 168 monitored models (20). On 2 November 2026 the API component shows 100 per cent over 86.7 measured days and the website 99.99 per cent, with one 8-minute degradation on 19 August 2026. The Atom feed's 50 newest items, 20 August to 30 October 2026, are all automated major-outage observations on single models, "not an announced incident", among them MiMo-V2.5 from 27 August to 4 September, Qwen3-TTS for 37 and 32 hours, Nemotron-3-Ultra for 25 hours and a run of short Nemotron-3.5-Lightning outages in early September. No written incident touched the API. Each observation is one model, partial under this run's rule, as together-inference's single-model monitor downtime was read, so minor only (20). The 8 October reading of 15 sat between checklist values. 200 concurrent requests per model, published (15). 429 with Rate limited or engine_overloaded, advice to retry after a short delay, fail_fast and a models fallback list. No Retry-After header or backoff figures were found (10 of 15). No SLA found. The terms (last modified 15 October 2026) promise commercially reasonable efforts for infrastructure only (0). Generally available (10). 20 + 20 + 15 + 10 + 0 + 10 = 75.