Amazon Bedrock, 서울 리전에서 지금 쓸 수 있는 모델
ap-northeast-2의 bedrock:ListFoundationModels·ListInferenceProfiles 응답을 매일 06:00 KST에 다시 조회해 자동 갱신합니다. 콘솔 카탈로그를 리전마다 뒤지지 않아도 되도록 — 서울에 있는 것, 없는 것, 바뀐 것을 한 페이지에.
마지막 확인 · 원본 데이터 JSON · RSS· 생성 스크립트
- 31서울 모델 (base 30)
- 5On-demand 호출 가능
- 25추론 프로필(CRIS) 전용
- 26%us-east-1 대비 (27/104)
- 86서울에 없는 모델
서울에서 쓸 수 있는 모델 31
| Provider | Model | Model ID | In → Out | Stream | Access | Since |
|---|---|---|---|---|---|---|
| Amazon | Nova 2 Lite | amazon.nova-2-lite-v1:0 | Image + Text + Video → Text | ✓ | CRIS global.amazon.nova-2-lite-v1:0 | 2025-12-02 |
| Amazon | Nova Lite | amazon.nova-lite-v1:0 | Image + Text + Video → Text | ✓ | CRIS apac.amazon.nova-lite-v1:0 | 2024-12-03 |
| Amazon | Nova Micro | amazon.nova-micro-v1:0 | Text → Text | ✓ | CRIS apac.amazon.nova-micro-v1:0 | 2024-12-03 |
| Amazon | Nova Pro | amazon.nova-pro-v1:0 | Image + Text + Video → Text | ✓ | CRIS apac.amazon.nova-pro-v1:0 | 2024-12-03 |
| Amazon | Titan Embeddings G2 - Text | amazon.titan-embed-text-v2:0 | Text → Embedding | — | On-demand | 2024-04-30 |
| Anthropic | Claude 3 Haiku | anthropic.claude-3-haiku-20240307-v1:0 | Image + Text → Text | ✓ | On-demandCRIS apac.anthropic.claude-3-haiku-20240307-v1:0 | 2024-03-13 |
| Anthropic | Claude 3 Haiku200k | anthropic.claude-3-haiku-20240307-v1:0:200k | Image + Text → Text | ✓ | — | 2024-03-13 |
| Anthropic | Claude 3.5 Sonnet | anthropic.claude-3-5-sonnet-20240620-v1:0 | Image + Text → Text | ✓ | On-demandCRIS apac.anthropic.claude-3-5-sonnet-20240620-v1:0 | 2024-06-20 |
| Anthropic | Claude 3.5 Sonnet v2 | anthropic.claude-3-5-sonnet-20241022-v2:0 | Image + Text → Text | ✓ | CRIS apac.anthropic.claude-3-5-sonnet-20241022-v2:0 | 2024-10-22 |
| Anthropic | Claude Fable 5 | anthropic.claude-fable-5 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-fable-5 | 2026-06-09 |
| Anthropic | Claude Fable 5.1 | anthropic.claude-fable-5-1 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-fable-5-1 | 2026-08-31 |
| Anthropic | Claude Haiku 4.5 | anthropic.claude-haiku-4-5-20251001-v1:0 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-haiku-4-5-20251001-v1:0 | 2025-10-15 |
| Anthropic | Claude Opus 4.5 | anthropic.claude-opus-4-5-20251101-v1:0 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-opus-4-5-20251101-v1:0 | 2025-11-24 |
| Anthropic | Claude Opus 4.6 | anthropic.claude-opus-4-6-v1 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-opus-4-6-v1 | 2026-02-05 |
| Anthropic | Claude Opus 4.7 | anthropic.claude-opus-4-7 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-opus-4-7 | 2026-04-16 |
| Anthropic | Claude Opus 4.8 | anthropic.claude-opus-4-8 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-opus-4-8 | 2026-05-28 |
| Anthropic | Claude Opus 5 | anthropic.claude-opus-5 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-opus-5 | 2026-07-23 |
| Anthropic | Claude Sonnet 4 | anthropic.claude-sonnet-4-20250514-v1:0 | Image + Text → Text | ✓ | CRIS apac.anthropic.claude-sonnet-4-20250514-v1:0 | 2025-05-22 |
| Anthropic | Claude Sonnet 4.5 | anthropic.claude-sonnet-4-5-20250929-v1:0 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-sonnet-4-5-20250929-v1:0 | 2025-09-29 |
| Anthropic | Claude Sonnet 4.6 | anthropic.claude-sonnet-4-6 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-sonnet-4-6 | 2026-02-17 |
| Anthropic | Claude Sonnet 5 | anthropic.claude-sonnet-5 | Image + Text → Text | ✓ | CRIS global.anthropic.claude-sonnet-5 | 2026-06-30 |
| Cohere | Embed v4 | cohere.embed-v4:0 | Image + Text → Embedding | — | CRIS global.cohere.embed-v4:0 | 2025-10-02 |
| Moonshot AI | Kimi K3 | moonshotai.kimi-k3 | Image + Text → Text | ✓ | CRIS global.moonshotai.kimi-k3 | 2026-08-27 |
| OpenAI | GPT-5.6 Luna | openai.gpt-5.6-luna | Image + Text → Text | ✓ | CRIS global.openai.gpt-5.6-luna | 2026-08-13 |
| OpenAI | GPT-5.6 Sol | openai.gpt-5.6-sol | Image + Text → Text | ✓ | CRIS global.openai.gpt-5.6-sol | 2026-08-13 |
| OpenAI | GPT-5.6 Terra | openai.gpt-5.6-terra | Image + Text → Text | ✓ | CRIS global.openai.gpt-5.6-terra | 2026-08-13 |
| OpenAI | GPT-6 Astra | openai.gpt-6-astra | Image + Text → Text | ✓ | CRIS global.openai.gpt-6-astra | 2026-09-08 |
| TwelveLabs | Marengo Embed 3.0 | twelvelabs.marengo-embed-3-0-v1:0 | Image + Speech + Text + Video → Embedding | — | On-demand | 2025-10-28 |
| TwelveLabs | Marengo Embed v2.7 | twelvelabs.marengo-embed-2-7-v1:0 | Image + Speech + Text + Video → Embedding | — | CRIS apac.twelvelabs.marengo-embed-2-7-v1:0 | 2025-07-15 |
| TwelveLabs | Pegasus v1.2 | twelvelabs.pegasus-1-2-v1:0 | Text + Video → Text | ✓ | On-demandCRIS apac.twelvelabs.pegasus-1-2-v1:0global.twelvelabs.pegasus-1-2-v1:0 | 2025-12-16 |
| xAI | Grok 4.6 | xai.grok-4.6 | Image + Text → Text | ✓ | CRIS global.xai.grok-4.6 | 2026-08-19 |
Access — On-demand는 modelId로 직접 호출, CRIS는 표시된 추론 프로필 ID(apac.*/global.*)로만 호출 가능(cross-region inference). 접근 경로가 없는 항목은 카탈로그에만 있는 컨텍스트 변형(provisioned 전용)입니다. 카탈로그 기준이라 계정별 모델 액세스 승인·쿼터·요금은 반영하지 않습니다.
서울에 없는 모델 86
추적 리전 중 하나 이상에서 제공되지만 서울엔 없는 base 모델(컨텍스트 변형은 하나로 합침). 리전 약어: use1 버지니아 (115) · usw2 오레곤 (109) · use2 오하이오 (86) · apne1 도쿄 (69)
- Stability AI16Stable Diffusion 3.5 Largeusw2Stable Image Conservative Upscaleuse1 usw2 use2Stable Image Control Sketchuse1 usw2 use2Stable Image Control Structureuse1 usw2 use2Stable Image Core 1.0usw2Stable Image Creative Upscaleuse1 usw2 use2Stable Image Erase Objectuse1 usw2 use2Stable Image Fast Upscaleuse1 usw2 use2Stable Image Inpaintuse1 usw2 use2Stable Image Outpaintuse1 usw2 use2Stable Image Remove Backgrounduse1 usw2 use2Stable Image Search and Recoloruse1 usw2 use2Stable Image Search and Replaceuse1 usw2 use2Stable Image Style Guideuse1 usw2 use2Stable Image Style Transferuse1 usw2 use2Stable Image Ultra 1.0usw2
- Mistral AI14Devstral 2 123Buse1 usw2 use2 apne1Magistral Small 2509use1 usw2 use2 apne1Ministral 14B 3.0use1 usw2 use2 apne1Ministral 3 8Buse1 usw2 use2 apne1Ministral 3Buse1 usw2 use2 apne1Mistral 7B Instructuse1 usw2Mistral Large (24.02)use1 usw2Mistral Large (24.07)usw2Mistral Large 3use1 usw2 use2 apne1Mistral Small (24.02)use1Mixtral 8x7B Instructuse1 usw2Pixtral Large (25.02)use1 usw2 use2Voxtral Mini 3B 2507use1 usw2 use2 apne1Voxtral Small 24B 2507use1 usw2 use2 apne1
- Amazon10Amazon Nova Multimodal Embeddingsuse1Nova 2 Sonicuse1 usw2 apne1Nova Canvasuse1 apne1Nova Reeluse1 apne1Nova Reeluse1Rerank 1.0usw2 apne1Titan Embeddings G1 - Textuse1 usw2 apne1Titan Multimodal Embeddings G1use1 usw2Titan Multimodal Embeddings G1use1 usw2Titan Text Embeddings v2use1 usw2
- Meta7Llama 3 70B Instructuse1 usw2Llama 3 8B Instructuse1 usw2Llama 3.1 70B Instructuse1 usw2 use2Llama 3.1 8B Instructuse1 usw2 use2Llama 3.3 70B Instructuse1 usw2 use2Llama 4 Maverick 17B Instructuse1 usw2 use2Llama 4 Scout 17B Instructuse1 usw2 use2
- Qwen7Qwen3 235B A22B 2507usw2 use2 apne1Qwen3 32B (dense)use1 usw2 use2 apne1Qwen3 Coder 480B A35B Instructusw2 use2 apne1Qwen3 Coder Nextuse1Qwen3 Next 80B A3Buse1 usw2 use2 apne1Qwen3 VL 235B A22Buse1 usw2 use2 apne1Qwen3-Coder-30B-A3B-Instructuse1 usw2 use2 apne1
- NVIDIA4Nemotron Nano 3 30Buse1 usw2 use2 apne1NVIDIA Nemotron 3 Super 120B A12Buse1 usw2 use2 apne1NVIDIA Nemotron Nano 12B v2 VL BF16use1 usw2 use2 apne1NVIDIA Nemotron Nano 9B v2use1 usw2 use2 apne1
- OpenAI4GPT OSS Safeguard 120Buse1 usw2 use2 apne1GPT OSS Safeguard 20Buse1 usw2 use2 apne1gpt-oss-120buse1 usw2 use2 apne1gpt-oss-20buse1 usw2 use2 apne1
- Cohere3Embed Englishuse1 usw2 apne1Embed Multilingualuse1 usw2 apne1Rerank 3.5use1 usw2 apne1
- DeepSeek3DeepSeek V3.2use1 usw2 use2 apne1DeepSeek-R1use1 usw2 use2DeepSeek-V3.1usw2 use2 apne1
- Google3Gemma 3 12B ITuse1 usw2 use2 apne1Gemma 3 27B PTuse1 usw2 use2 apne1Gemma 3 4B ITuse1 usw2 use2 apne1
- MiniMax3MiniMax M2use1 usw2 use2 apne1MiniMax M2.1use1 usw2 use2 apne1MiniMax M2.5use1 usw2 use2 apne1
- Writer3Palmyra X4use1 usw2 use2Palmyra X5use1 usw2 use2Writer Palmyra Vision 7Buse1 usw2 use2
- Z.AI3GLM 4.7use1 usw2 use2 apne1GLM 4.7 Flashuse1 usw2 use2 apne1GLM 5use1 usw2 use2 apne1
- AI21 Labs2Jamba 1.5 Largeuse1Jamba 1.5 Miniuse1
- Moonshot AI2Kimi K2 Thinkinguse1 usw2 use2 apne1Kimi K2.5use1 usw2 use2 apne1
- Anthropic1Claude Opus 4.1use1 usw2 use2
- Luma AI1Ray v2usw2
서울에서 쓸 수 있는 추론 프로필 30
시스템 정의 cross-region inference 프로필. apac.*은 APAC 리전 묶음으로 라우팅, global.*은 글로벌 엔드포인트(요청은 서울에서 나가지만 처리 리전은 AWS가 선택). 데이터 상주 요건이 있으면 라우팅 대상 리전을 확인하세요.
apac.*10
apac.amazon.nova-lite-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-southeast-1 ap-southeast-2apac.amazon.nova-micro-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-southeast-1 ap-southeast-2apac.amazon.nova-pro-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-southeast-1 ap-southeast-2apac.anthropic.claude-3-5-sonnet-20240620-v1:0ap-northeast-1 ap-northeast-2 ap-south-1 ap-southeast-1 ap-southeast-2apac.anthropic.claude-3-5-sonnet-20241022-v2:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-southeast-1 ap-southeast-2apac.anthropic.claude-3-haiku-20240307-v1:0ap-northeast-1 ap-northeast-2 ap-south-1 ap-southeast-1 ap-southeast-2apac.anthropic.claude-3-sonnet-20240229-v1:0ap-northeast-1 ap-northeast-2 ap-south-1 ap-southeast-1 ap-southeast-2apac.anthropic.claude-sonnet-4-20250514-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-south-2 ap-southeast-1 ap-southeast-2 ap-southeast-4apac.twelvelabs.marengo-embed-2-7-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-south-2 ap-southeast-1 ap-southeast-2 ap-southeast-3 ap-southeast-4apac.twelvelabs.pegasus-1-2-v1:0ap-northeast-1 ap-northeast-2 ap-northeast-3 ap-south-1 ap-south-2 ap-southeast-1 ap-southeast-2 ap-southeast-4
global.*20
global.amazon.nova-2-lite-v1:0global endpointglobal.anthropic.claude-fable-5global endpointglobal.anthropic.claude-fable-5-1global endpointglobal.anthropic.claude-haiku-4-5-20251001-v1:0global endpointglobal.anthropic.claude-opus-4-5-20251101-v1:0global endpointglobal.anthropic.claude-opus-4-6-v1global endpointglobal.anthropic.claude-opus-4-7global endpointglobal.anthropic.claude-opus-4-8global endpointglobal.anthropic.claude-opus-5global endpointglobal.anthropic.claude-sonnet-4-5-20250929-v1:0global endpointglobal.anthropic.claude-sonnet-4-6global endpointglobal.anthropic.claude-sonnet-5global endpointglobal.cohere.embed-v4:0global endpointglobal.moonshotai.kimi-k3global endpointglobal.openai.gpt-5.6-lunaglobal endpointglobal.openai.gpt-5.6-solglobal endpointglobal.openai.gpt-5.6-terraglobal endpointglobal.openai.gpt-6-astraglobal endpointglobal.twelvelabs.pegasus-1-2-v1:0global endpointglobal.xai.grok-4.6global endpoint
변경 이력 5
매일 이전 스냅샷과 비교해 추가·제거·속성 변경(status, inference type, modality, EOL)을 기록합니다. RSS로 구독하면 서울에 새 모델이 열리는 날 알 수 있습니다.
- ap-northeast-2첫 스냅샷 — 31개 모델
- us-east-1첫 스냅샷 — 115개 모델
- us-west-2첫 스냅샷 — 109개 모델
- us-east-2첫 스냅샷 — 86개 모델
- ap-northeast-1첫 스냅샷 — 69개 모델
데이터로 쓰기
페이지는 껍데기고 산출물은 아래 파일입니다. 그대로 가져다 쓰세요 (CORS 허용, 인증 없음, 매일 갱신). 파생 데이터는 CC0.
- seoul.json/data/bedrock/seoul.json — 서울 모델 + 접근 경로 + 부재 목록 + 통계
- models.json/data/bedrock/models.json — 추적 리전 5곳 전체 원본 정규화
- profiles.json/data/bedrock/profiles.json — 서울 추론 프로필과 라우팅 리전
- changes.json/data/bedrock/changes.json — 변경 이력 (최신순)
- feed.xml/data/bedrock/feed.xml — 변경 이력 RSS 2.0
- meta.json/data/bedrock/meta.json — 마지막 확인 시각·소스
# 서울에서 on-demand로 바로 부를 수 있는 모델 ID
curl -s https://hub.sean-chloe.com/data/bedrock/seoul.json \
| jq -r '.models[] | select(.access | index("on-demand")) | .id'
# Claude 계열이 서울에서 어떤 프로필로 열려 있나
curl -s https://hub.sean-chloe.com/data/bedrock/seoul.json \
| jq -r '.models[] | select(.provider=="Anthropic") | "\(.name)\t\(.profiles|join(","))"'소스: AWS Bedrock 컨트롤플레인 API 공개 카탈로그 응답. 프롬프트 캐싱·툴 사용·구조화 출력 같은 기능 지원 여부는 API가 돌려주지 않아 포함하지 않았습니다(문서 기반 오버레이는 다음 단계). 틀린 점·추가 리전 요청은 GitHub 이슈로.