Sulat.com
AI models
$10 off the fastest DeepSeek V4.1 Flash, Kimi K3 and GLM 5.3 from Synthetic
Amazon Bedrock logo

Model details

Claude Sonnet 4.5 (JP)

Claude Sonnet 4.5 (JP) sits in the middle of Anthropic's Claude Sonnet family and is offered through Amazon Bedrock, giving Japanese-region customers a managed route to the model. It accepts text, images, and PDF documents as input and produces text, with built-in support for tool calling, structured output, and configurable temperature. Third-party commentary around its Bedrock launch frames it as a general-purpose assistant positioned for productivity and coding tasks, while independent fit scoring highlights particularly strong agent-style usage thanks to the combination of tool calling, structured output, and reasoning features.

The model is designed for practical, production-oriented workloads rather than open experimentation, since it is distributed as a hosted service without open weights. Independent profile scoring rates its coding ability and vision handling favorably, reflecting benchmarks and capability coverage tied to image input and reasoning. Pricing sits at $the listed price per million input tokens and $the listed price per million output tokens, with prompt-caching available at reduced rates, making it a middle-tier option for teams that need a balance of capability and cost on Bedrock rather than the cheapest or the largest model in the lineup.

Amazon Bedrockjp.anthropic.claude-sonnet-4-5-20250929-v1:0claude-sonnet

Quick Info

Powered by
Provider
Amazon Bedrock
Model key
jp.anthropic.claude-sonnet-4-5-20250929-v1:0
Release date
Sep 29, 2025
Last updated
Sep 29, 2025
Knowledge cutoff
2025-07-31
Input modalities
Output modalities
Capabilities

Cost

Input token cost
$3.30
Output token cost
$16.50

Limits

Output tokens
64,000 tokens
Context window
200,000 tokens

Latest news about Claude Sonnet 4.5 (JP)

Amazon Bedrock

CoverageBenchmark

Claude Sonnet 4.5 is tracked on BenchLM with a capability score of 53.7 out of 100 and an overall rank of 108 out of 232 models in their catalog. Its strongest eligible category is Coding, where it ranks 67, placing it in the 64th percentile for software development and code generation among its comparison cohort. The Eleven benchmark rows are published for Claude Sonnet 4.5 on the BenchLM record, with several tracked benchmark slots left empty and independent runtime speed not yet measured. Agentic capability ranks at the 34th percentile and Knowledge at the 56th percentile, while Reasoning, Math, Multilingual, Multimodal, and Inst

Videos about Claude Sonnet 4.5 (JP)

More models around Claude Sonnet 4.5 (JP)