Amazon Bedrock vs Together AI 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.

Together AI Inference 70

Project-scoped, revocable keys with optional expiry and rotation guidance (30). Project boundaries but keys can access all project resources and lack per-key spend limits (10 of 20). No inference prompt-injection mitigation established (0). Per-key cost and usage attribution (15). Published security.txt with a future expiry and HackerOne disclosure route (15 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.

Together AI Inference 65

Graded as listed, on the tool lines. Changelog entries on 22 and 23 October 2026 (Python SDK 3.0, TypeScript SDK 1.0, MiniMax M3.1 Flash on serverless and post-training) (30). 43 dated changelog days and 16 Python SDK tags from 3 August, v2.33.1 to v3.0.0 (20). Support responsiveness not tested (0). Official Python and TypeScript SDKs, both given new majors on 22 October (15). Dependency and CI health not audited (0).

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).

Together AI Inference 74

Proprietary service with published terms and Apache-2.0 SDK (15). Docs distinguish default content storage, training opt-in, organisation ZDR and passthrough models, but default content retention has no fixed duration in the reviewed pages (25 of 30). Public lifecycle policy gives a two-week migration window for breaking model upgrades and 2 or 3 weeks' notice for serverless removals, though removals on 19, 21, 25 and 27 August 2026 were posted only on the day (15 of 20). North American hosting and enterprise residency options are described; full subprocessor locations were not established (10 of 20).

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).

Together AI Inference 20

No machine payment protocol found in the reviewed billing and inference docs (0). Public token and capacity prices (20). No free trial and a $15 minimum first credit purchase (0). Browser account and payment setup required (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).

Together AI Inference 85

Compact model API with bounded output (25). Token controls, streaming, structured outputs and batch options (20). Actionable error documentation (20). Transient retry guidance but no general inference deduplication guarantee found (5 of 20). Official Python and TypeScript SDKs and defaults (15).

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).

Together AI Inference 75

Graded as listed, on the hosted API lines. Public Better Stack status page at status.together.ai with a component for each serverless model plus the website and playground (20). The page shows 90 daily ticks per component, read on 1 November 2026 for 4 August to 1 November, and no incident was posted in that time (the feed's newest items are maintenance windows in April and July 2026). Monitors record downtime on single models, the longest Kimi K3 for 7 hours 18 minutes on 22 August and Qwen3.8-2.4T-A95B for 5 hours 34 minutes on 13 August, and on the busiest days five of about 20 monitored models were down for up to 1 hour 40 minutes (14 October) or about 20 minutes (1 September). Each hit one model or a subset of the catalogue, so the record reads as minor incidents only (20). The current serverless page gives no numeric general request ceiling (0). Explicit 429 and 503 backoff guidance (15). Provisioned throughput has a published 99% eligible-request commitment (10). Serverless inference is generally available (10). 20 + 20 + 0 + 15 + 10 + 10 = 75.