OpenAI compatible API · Attested · Public status

Amazon Bedrock performance

Name: Amazon Bedrock TrustedRouter performance measurements
Creator: TrustedRouter
License: https://www.apache.org/licenses/LICENSE-2.0

Review measured TTFT, effective throughput, uptime, and sampled model routes for Amazon Bedrock on TrustedRouter using metadata-only production probes.

Verify gateway

Onebase URL to migrate

100sof models and routes

0prompt or output logs. Always.

Amazon Bedrockamazon-bedrock

190 samples

Provider overview

Continuously sampled provider performance. TrustedRouter reports unsupported route and probe-configuration rows separately from provider downtime. Prompt and output content is not stored.

p50 TTFT	2101 ms
p95 TTFT	6251 ms
Effective throughput	80 tok/s n=1
Uptime	100.00%

Measured model routes

Model	p50 TTFT	Effective throughput	Uptime	Config excluded	Availability samples
anthropic/claude-opus-4.8	1816 ms	80 tok/s n=1	100.00%	—	21
anthropic/claude-sonnet-5.5	2033 ms	—	100.00%	—	47
anthropic/claude-opus-5.5	2101 ms	—	100.00%	—	39
anthropic/claude-sonnet-5	2139 ms	—	100.00%	—	22
anthropic/claude-haiku-5.5	2178 ms	—	100.00%	—	32
writer/palmyra-vision-7b	655 ms	—	100.00%	—	1
openai/gpt-oss-safeguard-120b	777 ms	—	100.00%	—	1
mistralai/ministral-3-8b-instruct	789 ms	—	100.00%	—	1
nvidia/nemotron-nano-12b-v2	877 ms	—	100.00%	—	1
deepseek/deepseek-v3.1	893 ms	—	100.00%	—	3
qwen/qwen3-coder-30b-a3b-instruct	912 ms	—	100.00%	—	1
mistralai/mistral-large-2512	990 ms	—	100.00%	—	3
minimax/minimax-m2.5	1052 ms	—	100.00%	—	2
google/gemma-3-12b-it	1152 ms	—	100.00%	—	2
minimax/minimax-m2	1292 ms	—	100.00%	—	2
mistralai/ministral-3-14b-instruct	1379 ms	—	100.00%	—	1
qwen/qwen3-coder-480b-a35b-instruct	1406 ms	—	100.00%	—	1
deepseek/deepseek-v3.2	1463 ms	—	100.00%	—	2
nvidia/nemotron-3-super-120b-a12b	1850 ms	—	100.00%	—	1
qwen/qwen3-coder-next	5285 ms	—	100.00%	—	1
qwen/qwen3-next-80b-a3b-instruct	11371 ms	—	100.00%	—	3
qwen/qwen3-235b-a22b-instruct-2507	12553 ms	—	100.00%	—	3