kimi-k2.5
AWS Bedrock/@us-east-2/chat10% off
Kimi K2.5 is an open-source, native multimodal agentic model built through continual pretraining on approximately 15 trillion mixed visual and text tokens atop Kimi-K2-Base. It seamlessly integrates vision and language understanding with advanced agentic capabilities, instant and thinking modes, as well as conversational and agentic paradigms.MoreLess
Which id to call
bedrock/kimi-k2.5@us-east-2This exact deployment on AWS Bedrock in us-east-2, with no routing and no failover. Send it as the model field.
Input /1M
$0.54
$0.60 list
AWS Bedrock
Output /1M
$2.70
$3.00 list
5.0x input
Context
128K
16K output
Added
Jan 2026
chat
Capabilities 4/8
Provider rates
What kimi-k2.5 costs
Provider prices per 1M tokens, updated August 24, 2026.moreless
These are the upstream provider rates. Pay as you go adds 5%, or 0% if you bring your own keys, and there is no per-request fee. Prompt caching and routing change what you pay against these rates, not the rates themselves.
This endpoint is discounted. List is $0.60 per 1M input and $3.00 per 1M output, and the rates below are what you pay. The discount applies to every request on this endpoint, with nothing to claim or enter.
Input /1M
$0.54
$0.60 list
Output /1M
$2.70
$3.00 list
Cache write /1M
-
Cache read /1M
-
What a workload costs
- 100K input + 10K output
- $0.0810
- 1M input + 100K output
- $0.81
- 10M input + 1M output
- $8.10
At the rates above. This endpoint does not offer prompt caching, so repeated context is billed as new input every time.
OpenAI compatible
Call it in three lines
Change the base url, use your Requesty key, set the model to bedrock/kimi-k2.5@us-east-2. Existing OpenAI SDK code needs no other edit.
123456789101112131415from openai import OpenAI client = OpenAI( api_key="YOUR_REQUESTY_API_KEY", base_url="https://router.requesty.ai/v1", ) response = client.chat.completions.create( model="bedrock/kimi-k2.5@us-east-2", messages=[ {"role": "user", "content": "Explain quantum computing in one paragraph."}, ], ) print(response.choices[0].message.content)
Reference
Specs and data terms
What the catalog reports for this deployment, and what the provider does with the traffic.
Released 2026-01-27
Benchmark scores
Benchmarks measure the model, not this endpoint, so they are the same wherever these weights are served.
Artificial Analysis Coding Index: a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.
Graduate-level physics, chemistry & biology questions designed to resist Googling.
Artificial Analysis Intelligence Index: a composite of multiple evaluations measuring overall model capability.
Scores from official model cards, Artificial Analysis and public leaderboards. They measure specific skills and do not capture every aspect of model quality, so test on your own workload.
Same provider
More from AWS Bedrock
Newest first, on the same provider and the same key.
Reference
kimi-k2.5 questions
How much does kimi-k2.5 cost?
kimi-k2.5 is priced at $0.54 per million input tokens and $2.70 per million output tokens when accessed via Requesty. Those figures include a 10% discount on this endpoint, off a list rate of $0.60 per million input tokens and $3.00 per million output tokens. Those are the upstream provider rates: pay as you go adds 5% on top, or 0% if you bring your own provider keys.
What is the context window of kimi-k2.5?
kimi-k2.5 has a context window of 128K tokens, with a maximum output of 16K tokens per response. That's roughly 171 words of input you can fit in a single prompt.
How does kimi-k2.5 perform on benchmarks?
kimi-k2.5 scores 95.9% on τ²-Bench, 87.9% on GPQA Diamond, 49.0% on SciCode. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can kimi-k2.5 do?
kimi-k2.5 supports vision input, tool calling, extended reasoning, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use kimi-k2.5 with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "bedrock/kimi-k2.5@us-east-2". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run kimi-k2.5 through Requesty?
Yes. kimi-k2.5 runs through Requesty's OpenAI-compatible API, served from AWS Bedrock in us-east-2. You do not host the model yourself: point base_url at Requesty, set the model to "bedrock/kimi-k2.5@us-east-2", and requests are routed to the upstream provider with automatic failover. The same key gives you 600+ other models too.
What region is this deployment?
This variant of kimi-k2.5 is deployed in us-east-2. Region-specific endpoints matter for data residency, latency to your users, and compliance requirements (GDPR, HIPAA). Other regions for the same model may be listed on the AWS Bedrock provider page.
Call kimi-k2.5 through one endpoint
One key for this endpoint and 600+ other models. No markup on provider prices, automatic failover, caching built in.