Amazon Bedrock vs Cerebras Inference

Amazon Bedrock 98

Model reading. IAM with SigV4, roles and STS session credentials, policies down to a model or inference-profile ARN, short-term API keys that expire within 12 hours, long-term keys with expiry and reset, and condition keys that can deny bearer tokens altogether (30). The security page says customer data isn't shared with model providers or used to improve the base models (20). The abuse detection page says Bedrock stores no inputs or outputs by default, and a per-Region none mode blocks models that require retention, but Claude Fable 5 and 5.1 keep all traffic up to 30 days and GPT-6 and GPT-5.x keep flagged traffic up to 30 days (13 of 15, a judgement call; Anthropic's own API got 12 for the same Fable exception without an account-level block). CloudTrail logs API calls, and model invocation logging can send prompts and responses to S3 or CloudWatch Logs (15). Bedrock is in ISO, SOC and CSA STAR Level 2 scope, HIPAA eligible and FedRAMP High in GovCloud (US-West) per its security page, with a vulnerability disclosure programme on HackerOne, and the aws.amazon.com security.txt, expired on 6 October, was renewed with Expires 2027-10-16 when read on 6 November (20). Up from 96 on 6 October for that reason.

Cerebras Inference 70

Revocable Bearer keys with rotation instructions (20). Paid-project boundaries, but no inference-only key permission found (10 of 20). GPT OSS documentation warns about undeclared tool calls and tells callers to validate them (10 of 15). Request and audit logs in the console (15). Trust centre lists SOC 2 Type II, with reports requiring access (15 of 20).

Amazon Bedrock 85

Closed service under the AWS Customer Agreement and Service Terms, with the SDKs and Smithy models under Apache-2.0 (18, as for S3). The data retention, abuse detection and security pages agree that Bedrock stores nothing by default, name the models that retain traffic and for how long, and say retained data isn't shared with providers. The retention page's default mode says AWS 'may retain the data for safety and abuse-prevention purposes' without a period, and a legacy provider_data_share mode is kept though sharing isn't supported (25 of 30). A written lifecycle policy with Legacy and EOL dates per model, but the legacy table's 34-day Legacy periods for five Claude models in some Regions contradict its own 6-month floor (15 of 20, down from 18). AWS's sub-processor list, per the 4 October S3 check, and per-Region, geographic and global inference options, with global routing able to process data in any commercial Region (18 of 20).

Cerebras Inference 60

SDK is Apache-2.0, but the hosted cloud terms were not readable in this session (5 of 30). Privacy policy states no inference input/output retention but gives no fixed operational-log retention period (20 of 30). Dated deprecations and API-version migration guidance without a general minimum notice commitment found (10 of 20). Trust centre lists subprocessors and infrastructure controls, but full data-location details were not established (10 of 20).

Amazon Bedrock 81

Model reading. The last model or API change is 30 October 2026 (OpenAI model API support and cross-Region inference in GovCloud), after structured outputs for Claude on both endpoints on 18 October (30). Models launched since 7 September 2026 get a 6-month or 45-day Legacy period, but the legacy page, which promises at least 6 months in Legacy, lists Claude 3 Haiku, 3 Sonnet, 3.5 Sonnet, 3.5 Sonnet v2 and 3.7 Sonnet as Legacy from 10 or 13 October with end of life on 13 November in several Regions, about 34 days (6 of 12, down from 10). Six models reached end of life in the 90 days to 6 November on five dates, Command R and R+ on 19 August, Claude 3 Haiku on 10 September, Nova Sonic on 14 September, Nova Canvas and Nova Reel on 30 September and Claude Sonnet 4 on 14 October (3 of 8, down from 4). Dated document history several times a month, What's New and re:Post, with direct support on a paid plan (12 of 15). SDK issue replies not sampled (5 of 10). Current official SDKs, boto3 1.43.125 on 5 November and @aws-sdk/client-bedrock-runtime 3.1163.0 (15). Supported runtimes stated, Python 3.10 or later and Node 20 or later (10). Down from 86 on 6 October.

Cerebras Inference 65

Dated changelog entry on 28 October 2026, adding five models to the Dedicated catalogue and documenting image inputs in tool messages and new reasoning controls (30). Seven dated changelog entries in the 90 days from 3 August, on 28 and 31 August, two on 3 September, 22 September, 9 October and 28 October (20). Support responsiveness not established (0). Official Python and TypeScript SDKs, both last released as 1.91.0 on 8 July 2026 (15). Runtime dependency health and CI were not audited (0).

Amazon Bedrock 80

AWS Health Dashboard with per-service, per-Region feeds and a public history file (20). The history lists Amazon Bedrock as disrupted in ME-CENTRAL-1 (UAE) and ME-SOUTH-1 (Bahrain) since March 2026, where AWS said on 15 September 2026 (still inside the 90-day window on 6 November) that it can't restore resources held only there. No other Bedrock event appears in the history in the 90 days to 6 November; an EU-SOUTH-2 packet-loss event on 4 October named Bedrock AgentCore, not Bedrock. This counts as one major event (10). StatusGator's third-party mirror shows 'Increased Error Rates' warnings on 24, 26 and 27 August and 10 September that we couldn't find in AWS's own history, so we didn't score them. Per-model token quotas with default values in the AWS General Reference, such as 30,000,000 tokens a minute for Opus 5.5 global inference (15). The error guide gives exponential backoff with jitter for 429, 503 and 529 and says to honour Retry-After, and the SDKs retry ModelNotReadyException up to five times (15). SLA of 99.9 per cent a month per Region for the model APIs, last updated 4 October 2023 (10). Converse and the current Claude, GPT-6 and Nova 2 Lite models are GA (10).

Cerebras Inference 70

Graded as listed, on the hosted API lines. Public component status history at status.cerebras.ai, with components for GPT-OSS-120B, Qwen-3.8-27b and the Developer Console (20). The history feed, read on 1 November 2026, runs back to November 2025, so the whole 90 days from 3 August were read. On 25 September 2026 Cerebras posted that parts of the service might be unavailable from 10:54 UTC, services were back at 13:41 UTC and the incident was resolved at 17:14 UTC, marked major and affecting both public model components. That is about 2 hours 47 minutes with the whole shared catalogue affected, one major. A 15-minute partial degradation of Gemma 4 31B on 27 August is minor, and the 30 August firewall maintenance was announced on 25 August (10). Numeric rate-limit tables, although the trial request limit is 5 a minute on the rate-limits page and 30 a minute on the GPT OSS page, re-read on 1 November (15). SDK error classes and bounded retries with backoff (15). No public shared-inference SLA established (0). Shared inference is generally available (10).

Amazon Bedrock 30

No x402, MPP or L402 (0). Per-token prices per model, tier and Region on the public pricing page, filled from AWS's public price feed (20). New accounts get up to $200 in Free Tier credits and the Free Tier page lists Bedrock among eligible services, and the Free Tier FAQ says most new customers don't need a payment method, but AWS may still ask for one, so half, as for S3 (10). A person signs up for AWS in a browser and sets up IAM before any key exists (0).

Cerebras Inference 20

No machine payment protocol found in the reviewed API and billing docs (0). Public token prices (20). The $5 trial requires a verified payment method (0). Account creation and payment setup require a person (0).

Amazon Bedrock 92

Model reading of the checklist (tool use, structured output, caching, context, batch, SDKs, errors). Converse takes tools with toolChoice auto, any or a named tool and strict: true, but forced choice depends on the model behind it (17 of 20). Structured outputs on supported models only (13 of 15). Explicit and implicit prompt caching, cache reads at $0.20 per million on Opus 5.5, 0.05x input, with 5-minute and 1-hour TTLs (15). 1M-token context on Opus 5.5, Sonnet 5.5 and Nova 2 Lite, a judgement call since Bedrock has no default model (15). Batch at half price on supported models, but the pricing page shows none for Opus 5.5 or Sonnet 5.5, while Flex is half price on Nova (7 of 10). AWS SDKs in every major language plus token generators for Python, JavaScript and Java (10). Typed errors with recovery advice per code (15).

Cerebras Inference 85

Model API with bounded output and compact request shape (25). Output token controls, streaming and structured outputs (20). Documented error classes and fixes (20). Bounded transient retries but no request-deduplication guarantee established (5 of 20). Defaults and official Python and TypeScript SDKs (15).