---
title: "LLM API SLA compared: Azure, Bedrock, Gemini, OpenAI"
description: "What each LLM API provider commits to in writing: uptime percentage, what counts as downtime, service credits and latency targets, read from the SLA pages."
url: https://broadrouter.com/blog/llm-api-sla-uptime-compared/
lang: en
updated: 2026-11-15
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# LLM API SLAs compared: what Azure, Bedrock, Google and OpenAI promise

By the BroadNet Inference team · 7 min read · Published 2026-11-15 · Updated 2026-11-16

Most LLM APIs promise no uptime on pay-as-you-go. Written SLAs come with cloud contracts or reserved capacity: Azure Foundry Models and Amazon Bedrock credit you below 99.9% monthly uptime, Google's Gemini online inference below 99.5%, and OpenAI's Scale Tier offers 99.9% to Enterprise buyers. Anthropic's commercial terms and BroadNet's self-serve terms have none. The remedy is a bill credit.

## Key takeaways

-   Azure Foundry Models (which include all Azure OpenAI models) and Amazon Bedrock pay service credits below 99.9% monthly uptime. Google's Gemini online inference SLA starts at 99.5%.
-   OpenAI's 99.9% uptime SLA is part of Scale Tier, an Enterprise offer bought in token units for at least 30 days. We found no uptime SLA for standard pay-as-you-go.
-   Anthropic's Commercial Terms provide the service "as is" and "as available". BroadNet's self-serve terms have "no availability commitment".
-   Downtime usually means server errors (5XX or 500) only. Rate limits, slow answers and bad output are not downtime unless a separate latency SLA applies.
-   Credits are capped discounts on future bills, so design for failure anyway: retries, timeouts and a fallback provider.

**On this page**

1.  What does an LLM API SLA actually promise?
2.  Which LLM APIs publish an SLA?
3.  How do the hyperscaler SLAs differ?
4.  What about OpenAI and Anthropic directly?
5.  What does an SLA not buy you?
6.  How do you build on an API with no SLA?
7.  Where BroadNet Inference fits

Search “LLM API SLA” and you get tables that disagree with each other. This post skips the tables and reads the source: each provider’s own SLA or terms page, fetched on 15 November 2026. For each one we note the uptime target, what counts as downtime, the service credit, any latency target and which tier is covered. BroadRouter is BroadNet’s website, so BroadNet is in the list, and it is one of the providers without an SLA on self-serve.

## What does an LLM API SLA actually promise?

An [SLA](https://broadrouter.com/glossary/sla/) is a written target with a remedy. It does not make the API stay up. It says what happens on your bill when it doesn’t. Every SLA below has the same four parts:

1.  **A target.** Usually a monthly uptime percentage per region, project or resource.
2.  **A definition of downtime.** Almost always server errors only: 500 or 5XX responses to valid requests. A 429 [rate limit](https://broadrouter.com/glossary/rate-limit/) is not downtime. Neither is a slow answer or a wrong answer.
3.  **A remedy.** A percentage of that month’s bill for the affected service, credited to future bills and capped.
4.  **A claim process.** You ask for the credit, with your own logs, inside a deadline.

The percentages translate into minutes like this, for a 30-day month of 43,200 minutes:

| Monthly uptime | Downtime allowed |
| --- | --- |
| 99.9% | about 43 minutes |
| 99.5% | 216 minutes |
| 99.0% | 432 minutes |
| 95.0% | 36 hours |

## Which LLM APIs publish an SLA?

All rows read from the provider’s own page on 15 November 2026. “Not found” means the page we checked states no SLA; it does not rule out a private contract.

| Provider and service | Uptime target | Counts as downtime | Credits | Latency SLA | Who is covered |
| --- | --- | --- | --- | --- | --- |
| Microsoft Azure, Foundry Models (includes Azure OpenAI models) | 99.9% | Requests returning an error code, averaged per minute | 10% below 99.9%, 25% below 99% | Yes, for Provisioned Managed deployments and Priority Processing | Foundry Models sold by Azure; Developer tier excluded |
| Amazon Bedrock | 99.9% (credits start below it) | Requests that return a 500 error code, per 5-minute interval, per region | 10%, 25% below 99.0%, 100% below 95.0% | None in the SLA | Bedrock in each AWS region |
| Google, Gemini online inference API | 99.5% (95% for models with shorter availability) | Over 5% 5XX error rate for 5 or more consecutive minutes | 10%, 25% below 99.0%, 50% below 95.0% | Yes, under Provisioned Throughput on the global endpoint | `generateContent` and `streamGenerateContent` |
| OpenAI, Scale Tier | 99.9% | Not defined on the public page | Not stated on the public page | 99% of time above a per-model tokens-per-second floor | Enterprise customers, token units bought for at least 30 days |
| OpenAI, standard pay-as-you-go | Not found | – | – | – | – |
| Anthropic API | Not found; terms say “as is” and “as available” | – | – | – | – |
| Fireworks AI, Reserved Throughput | 99.9% uptime terms | Not stated on the pricing page | Not stated | Throughput SLA | Reserved capacity only; overage not covered |
| Together AI | Serverless: none stated. “Provisioned Throughput: Token-based capacity with SLAs” (terms not published on the page) | – | – | – | Ask sales for the SLA terms |
| BroadNet Inference, self-serve | None: “no availability commitment” | – | – | – | Service levels only in a separate written agreement |

## How do the hyperscaler SLAs differ?

**Microsoft Azure.** Microsoft’s Learn page says Foundry Models sold by Azure “include all Azure OpenAI models” and are “covered by Azure service-level agreements”. The consolidated SLA (October 2026 edition) defines Error Rate per minute and Uptime Percentage as 100% minus the average error rate. Credits are 10% below 99.9% and 25% below 99%. Credits apply only to Foundry Models sold by Azure, and “The Developer tier of the Foundry Models service is not covered by this SLA.” For Provisioned Managed deployments and Priority Processing, a second SLA measures tokens per second against a per-model target in the product docs, with a 10% credit if attainment falls below 99%.

**Amazon Bedrock.** AWS commits to “commercially reasonable efforts” to meet the uptime table. Availability is “the percentage of Requests processed by Amazon Bedrock that do not fail with Errors”, and an Error is “any Request that returns a 500 error code”. It is calculated per region, and intervals with no requests count as fully available. Exclusions include customer actions such as “scaling of provisioned throughput” and model software causing “repeated model crashes or an inoperable model”. The SLA has no latency target.

**Google Cloud.** The Gemini online inference SLA is the strictest about what counts. Downtime needs more than a 5% error rate of 5XX responses, lasting “five or more consecutive minutes”. Credits are capped at 50% of the month’s bill for the service, you must notify support within 30 days with log files, and the SLA is your “sole and exclusive remedy”. Grounding with Google Search, quota errors and deadlines shorter than the server default are excluded. Under Provisioned Throughput, a latency SLA covers some Gemini 2.5 models on the global endpoint, measured as p50 tokens per second per 5-minute interval. If you deploy your own open model on Vertex AI instead, the separate Vertex AI Platform SLA covers custom model online prediction at 99.5%, but only “for models deployed on 2 or more nodes”.

## What about OpenAI and Anthropic directly?

**OpenAI.** The Scale Tier page says: “Scale Tier traffic offers a 99.9% uptime SLA and prioritized compute.” Its table pairs that with a latency SLA per model, such as “99% > 50 tokens per second”, calculated as p50 request latency per 5 minutes. The page says “This offering is available to Enterprise customers”, that each token unit is bought “for a minimum of 30 days”, and that Scale Tier covers models released before GPT-5.6, with Reserved Tier for later models. We found no uptime SLA attached to standard pay-as-you-go use. (openai.com blocked our fetcher, so we read the page from the Internet Archive’s 7 November 2026 capture.)

**Anthropic.** The public Commercial Terms (effective 17 June 2025) contain no SLA or service credits. Section L.2 provides the services “AS IS” and “AS AVAILABLE” and disclaims that use “WILL BE UNINTERRUPTED”. Enterprise agreements may differ, but they are not public.

## What does an SLA not buy you?

-   **Uptime itself.** A credit is paid after the outage. Your users still saw it.
-   **Capacity.** Throttling (429) is outside every uptime definition above. Google explicitly excludes “quotas applied by the system”. See [how to handle 429 errors](https://broadrouter.com/blog/handle-llm-api-429-rate-limits/).
-   **Speed.** Only the provisioned or reserved tiers carry a tokens-per-second target. On standard traffic, a slow answer is a successful answer. [Latency metrics explained](https://broadrouter.com/blog/llm-latency-ttft-throughput/).
-   **Quality.** A 200 response with a wrong answer is not downtime anywhere.
-   **Much money.** Credits are a share of one service’s monthly bill, often capped, and you must claim them with your own logs.

## How do you build on an API with no SLA?

Most teams build on pay-as-you-go APIs with no SLA. That works if the application expects failure:

1.  **Set timeouts.** Stream long answers and set a client timeout per request, so a hung call fails fast. [Streaming guide](https://broadrouter.com/guides/streaming-chat-completions/).
2.  **Retry with backoff and jitter** on 429 and 5XX, and cap retries. [Errors and retries](https://broadrouter.com/guides/errors-and-retries/).
3.  **Keep a second provider configured.** With OpenAI-compatible APIs, a fallback is a second base URL, key and model ID. [How to fail over between LLM APIs](https://broadrouter.com/blog/llm-api-failover-fallback/).
4.  **Log request IDs, timestamps and status codes.** You need them for any SLA claim, and to see your real error rate.
5.  **Buy a contract where you need one.** If uptime is a hard requirement, use a provider tier with a written SLA, or a private deployment with agreed service levels. [What to plan before a private endpoint](https://broadrouter.com/blog/private-endpoint-planning/).

## Where BroadNet Inference fits

BroadRouter is BroadNet’s own website. BroadNet’s self-serve API, pay-as-you-go and plans alike, has no SLA. Its Terms of Service (§7.4) say: “We aim to keep the API available and publish what we measure, but self-serve has no availability commitment.” And: “Service levels exist only in a separate written agreement.” On this point Azure, AWS and Google offer more: a written uptime target with credits. A pay-as-you-go BroadNet key allows 30 requests a minute and 4 in flight. BroadNet’s own description of its serving setup is in [BroadNet’s GPU hardware and serving approach](https://broadrouter.com/blog/broadnet-gpu-fleet-and-serving/), which we report as BroadNet’s account, not as a commitment.

If your workload can run with retries and a fallback, BroadNet’s OpenAI-compatible endpoint is one more base URL to configure: `https://inference.broadnet.ai/v1`. If you need contracted service levels, ask BroadNet about a written agreement before you rely on one. [Read the API reference](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-llm-api-sla-uptime-compared&utm_content=body) or [create an account](https://inference.broadnet.ai/?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-llm-api-sla-uptime-compared&utm_content=body).

Free tools for this

-   [LLM API cost calculator](https://broadrouter.com/tools/llm-api-cost-calculator/) Price a workload on every model at once, with cache hits, reasoning and a flat-plan check.
-   [LLM token counter](https://broadrouter.com/tools/token-counter/) Exact token counts with each model's real tokenizer, with a token view and cost.
-   [Context window calculator](https://broadrouter.com/tools/context-window-calculator/) Will it fit? Window used per model, room left for output, cost to send.

[All free tools](https://broadrouter.com/tools/)

## Terms in this article

[SLA (service level agreement)](https://broadrouter.com/glossary/sla/)

An SLA (service level agreement) is a contractual commitment from a provider to a measurable level of service, usually a monthly uptime percentage, with a defined remedy, usually service credits, if it is missed. An SLA defines what counts as downtime, what is excluded and how to claim.

[Token](https://broadrouter.com/glossary/token/)

A token is the unit of text a language model reads and writes: a word, part of a word, a punctuation mark or a byte sequence produced by the model's tokenizer. LLM APIs count, limit and usually bill usage in tokens.

[Latency](https://broadrouter.com/glossary/latency/)

Latency, for a language model API, is the time a request takes from being sent to the final output token being received. It is the sum of time to first token and generation time, and is usually reported as percentiles such as p50 and p95 rather than as an average.

[Pay-as-you-go](https://broadrouter.com/glossary/pay-as-you-go/)

Pay-as-you-go is a pricing model in which you pay only for what you use, metered per unit such as tokens, requests or audio minutes, with no subscription or minimum commitment. Most LLM APIs bill this way, from a prepaid balance or a periodic invoice.

[Throughput](https://broadrouter.com/glossary/throughput/)

Throughput, in LLM serving, is the total amount of work a system completes per unit of time across all concurrent requests, usually measured as output tokens per second or requests per second. It differs from per-stream speed, which is how fast a single request receives its tokens.

[Rate limit](https://broadrouter.com/glossary/rate-limit/)

A rate limit is a cap an API places on how much a client can use it in a time window, such as requests per minute or tokens per minute. Requests over the limit are refused, usually with HTTP 429 Too Many Requests, until usage falls back under it.

[Tokens per second](https://broadrouter.com/glossary/tokens-per-second/)

Tokens per second (TPS) is the rate at which a language model generates output tokens, usually measured for one request after its first token arrives. It sets how fast a reply streams once it starts. Summed across all concurrent requests it becomes system throughput; the wait before the first token is measured separately.

[Base URL](https://broadrouter.com/glossary/base-url/)

A base URL is the fixed root address of an API, such as https://api.example.com/v1, that a client puts in front of every endpoint path, such as /chat/completions. In OpenAI-compatible SDKs, changing the base URL, together with the API key and model ID, points the same code at a different provider.

## Questions

### What is the Azure OpenAI SLA?

Azure OpenAI models are Foundry Models sold by Azure, which Microsoft says are covered by Azure SLAs. The October 2026 consolidated SLA gives Foundry Models a 10% service credit below 99.9% monthly uptime and 25% below 99%. The Developer tier is not covered. Provisioned Managed deployments and Priority Processing add a tokens-per-second latency SLA with a 10% credit below 99% attainment.

### Does the OpenAI API have an uptime SLA?

OpenAI publishes a 99.9% uptime SLA and a latency SLA for Scale Tier, which it describes as "available to Enterprise customers" and sells in token units for at least 30 days. For GPT-5.6 and later it points to Reserved Tier. We found no uptime SLA for standard pay-as-you-go API use.

### What is the Amazon Bedrock SLA?

AWS credits 10% of the region's Bedrock charges when monthly uptime falls below 99.9%, 25% below 99.0% and 100% below 95.0%. Availability is measured per region in 5-minute intervals, and an error is a request that returns a 500 error code.

### Does Anthropic offer an SLA for the Claude API?

Not in its public Commercial Terms. Those terms provide the services "as is" and "as available" and disclaim that use will be uninterrupted. Negotiated enterprise contracts are not public, so ask Anthropic, or use Claude through a cloud whose SLA covers it.

### How much downtime does 99.9% uptime allow?

About 43 minutes in a 30-day month: 0.1% of 43,200 minutes. 99.5% allows 216 minutes and 99% allows 432 minutes. How downtime is measured matters as much as the percentage.

### Does BroadNet Inference have an SLA?

Not on self-serve. BroadNet's Terms say "self-serve has no availability commitment" and "Service levels exist only in a separate written agreement", for example for a private endpoint.

## Keep reading

-   [How to handle LLM API 429 errors: backoff, jitter and limiters](https://broadrouter.com/blog/handle-llm-api-429-rate-limits/)
-   [How to choose an LLM API provider: a 20-point checklist](https://broadrouter.com/blog/choose-llm-api-provider-checklist/)
-   [A private endpoint starts with a clear workload](https://broadrouter.com/blog/private-endpoint-planning/)
-   [BroadNet's GPU hardware and serving approach, as BroadNet describes it](https://broadrouter.com/blog/broadnet-gpu-fleet-and-serving/)
-   [LLM latency: TTFT vs tokens per second vs end-to-end time](https://broadrouter.com/blog/llm-latency-ttft-throughput/)
-   [Artificial Analysis alternatives: where to compare LLMs in 2026](https://broadrouter.com/blog/artificial-analysis-alternatives/)
-   [Live, on request, research: read a model catalogue clearly](https://broadrouter.com/blog/understand-model-status/)
-   [Guide: BroadNet Inference error codes and when to retry](https://broadrouter.com/guides/errors-and-retries/)
-   [Guide: Stream chat completions and handle dropped connections](https://broadrouter.com/guides/streaming-chat-completions/)

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-llm-api-sla-uptime-compared&utm_content=page-cta) ·

**Sources**

-   [Microsoft, Service Level Agreements for Online Services, consolidated SLA October 2026 edition: Foundry Models section (checked 2026-11-15)](https://www.microsoft.com/licensing/docs/view/Service-Level-Agreements-SLA-for-Online-Services)
-   [Microsoft Learn, Foundry Models sold by Azure: "They include all Azure OpenAI models … covered by Azure service-level agreements" (updated 2026-10-29; checked 2026-11-15)](https://learn.microsoft.com/en-us/azure/ai-foundry/foundry-models/concepts/models-sold-directly-by-azure)
-   [AWS, Amazon Bedrock Service Level Agreement (last updated 4 October 2023; checked 2026-11-15)](https://aws.amazon.com/bedrock/sla/)
-   [Google Cloud, Gemini Online Inference API on Gemini Enterprise Agent Platform SLA (last modified 30 June 2026; checked 2026-11-15)](https://cloud.google.com/vertex-ai/generative-ai/sla)
-   [Google Cloud, Vertex AI Platform SLA (last modified 12 February 2026; checked 2026-11-15)](https://cloud.google.com/vertex-ai/sla)
-   [OpenAI, Scale Tier for API Customers (openai.com returned HTTP 403 to our fetcher; read from the Internet Archive capture of 7 November 2026, checked 2026-11-15)](https://web.archive.org/web/20261107081103/https://openai.com/api-scale-tier/)
-   [Anthropic, Commercial Terms of Service, effective 17 June 2025, sections I.3 and L.2 (checked 2026-11-15)](https://www.anthropic.com/legal/commercial-terms)
-   [Fireworks AI pricing, Reserved Throughput: "Throughput SLA and 99.9% uptime terms" (checked 2026-11-15)](https://fireworks.ai/pricing)
-   [Together AI pricing: SLAs offered with Provisioned Throughput only; no serverless SLA stated (checked 2026-11-15)](https://www.together.ai/pricing)
-   [BroadNet Terms of Service, §7.4 Availability (checked 2026-11-15)](https://www.broadnet.ai/terms)
