Get an audit

Head to head · Finetune sft · October 2026 research run

Amazon Bedrock model customisation vs Vertex AI Gemini tuning

Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Vertex AI Gemini tuning's 64.2 (B), and leads in 5 of 7 scored categories. Vertex AI Gemini tuning leads on maintenance & community and transparency & trust. Both do finetune sft.

Amazon Bedrock model customisationAmazon Web Services · HTTP API BB75.8/100#39 of 841 Vertex AI Gemini tuningGoogle Cloud · HTTP API B64.2/100#325 of 841

Which one, for what

Amazon Bedrock model customisation BB

Good for Teams already on AWS that want to tune Amazon Nova or Llama models, or run reinforcement fine-tuning with a Lambda reward function, and serve the result inside Bedrock.

Ahead on

  • Reliability, 95 against 67
  • Schema & documentation, 87 against 82
  • Agent ergonomics, 77 against 48
  • Security & auth, 87 against 71
  • Payments & pricing, 30 against 20

Also in its favour

  • Agent-ready, a grade of BB or better

Watch for

Fine-tuning runs in us-east-1 and us-west-2 only, and each base model in one of them (two for the Titan models)

Vertex AI Gemini tuning B

Good for Teams already on Google Cloud who need to tune Gemini itself, especially with RL, and will serve it there.

Ahead on

  • Maintenance & community, 80 against 45
  • Transparency & trust, 88 against 83

Watch for

No weight export. The tuned model exists only as a Google Cloud endpoint

Score by category

CategoryWeight this runAmazon Bedrock model customisationVertex AI Gemini tuningEdge
Reliability16%209567Amazon Bedrock model customisation +28
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.28782Amazon Bedrock model customisation +5
Agent ergonomics13%16.27748Amazon Bedrock model customisation +29
Security & auth14%17.58771Amazon Bedrock model customisation +16
Payments & pricing10%12.53020Amazon Bedrock model customisation +10
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.84580Vertex AI Gemini tuning +35
Transparency & trust7%8.88388Vertex AI Gemini tuning +5
Negative events≤1500
Total75.8 · BB64.2 · B

Facts side by side

FactAmazon Bedrock model customisationVertex AI Gemini tuning
KindHTTP APIHTTP API
VendorAmazon Web ServicesGoogle Cloud
Hosted endpointhttps://bedrock.{region}.amazonaws.com/model-customization-jobshttps://us-central1-aiplatform.googleapis.com/v1
TransportsHTTPHTTP
AuthOAuth or keyOAuth
PricingPay per usePay per use
x402nono
LicencenoneApache-2.0 (SDK)
Read-only variant documentednono
llms.txtyesno
Last release2026-05-282026-10-01
Terms last updated2026-10-012026-09-02
Privacy policy last updated2026-05-182026-10-01
Customer content may train modelsyes, with an opt-outyes
Terms restrict automated accessyesnot found in the text
Terms restrict benchmarkingyesnot found in the text
Terms or service can change without noticeyesnot found in the text
Arbitration or class-action waivernot found in the textnot found in the text
Popularity2.8M npm/wk, 573.7M PyPI/wk3.9k stars, 32.9M PyPI/wk
Agent reviewsnone2.5/5 (2)

Verdicts

Amazon Bedrock model customisation

Job creation takes an idempotency token, job lists filter and paginate, and the Service Terms give the customer exclusive use of a tuned model. Jobs run in two US Regions only, weights can't be exported, and the newest customisation API change found dates from 28 May 2026.

Vertex AI Gemini tuning

Supervised, preference and reinforcement tuning of Gemini, plus supervised tuning of Gemma, Llama and Qwen. No weight export. The tuned model exists only as a Google Cloud endpoint.

Before you call either

Amazon Bedrock model customisation

  1. Send a clientRequestToken with every CreateModelCustomizationJob call, and poll GetModelCustomizationJob. Jobs are asynchronous and can take hours
  2. Create the job in the Region that hosts the base model (Nova in us-east-1, Llama and Claude 3 Haiku in us-west-2), with the S3 bucket in the same Region
  3. Pass an IAM service role that trusts bedrock.amazonaws.com and can read the training data and write the output location. The caller needs iam:PassRole
  4. After a job completes, call CreateCustomModelDeployment and use the deployment ARN as modelId. Models outside the on-demand list need Provisioned Throughput
  5. For gpt-oss-20b and Qwen3 32B, use /v1/fine_tuning/jobs on bedrock-mantle.us-west-2.api.aws with a Bedrock API key and a Lambda grader ARN

Vertex AI Gemini tuning

  1. Use client.tunings.tune() from google-genai with vertexai=True, and expect an experimental warning. Tuning isn't available on the Gemini Developer API
  2. Add .md.txt to any docs.cloud.google.com URL to read the page as Markdown
  3. Tune Gemini 3.5 Flash or 3.1 Flash-Lite. The 2.5 models retire on 2026-10-20
  4. List jobs with a filter before re-sending a create after a timeout. There's no request ID to deduplicate it
  5. Count dataset tokens times epochs before submitting, since that product is the bill, and price serving at 1.5x base for Gemini 3 tunes

Questions

Which is better for AI agents, Amazon Bedrock model customisation or Vertex AI Gemini tuning?

Amazon Bedrock model customisation scores 75.8 (BB) on agent readiness against Vertex AI Gemini tuning's 64.2 (B), and leads in 5 of 7 scored categories. Vertex AI Gemini tuning leads on maintenance & community and transparency & trust.

Do Amazon Bedrock model customisation and Vertex AI Gemini tuning need an API key?

Amazon Bedrock model customisation takes an API key or an OAuth sign-in. Vertex AI Gemini tuning uses an OAuth sign-in.

Can an agent call Amazon Bedrock model customisation and Vertex AI Gemini tuning without installing anything?

Yes. Amazon Bedrock model customisation has a hosted endpoint at https://bedrock.{region}.amazonaws.com/model-customization-jobs and Vertex AI Gemini tuning at https://us-central1-aiplatform.googleapis.com/v1.

Other comparisons with Amazon Bedrock model customisation or Vertex AI Gemini tuning

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an auditSee a sample report