impulse

Kimi K3.1

Open weights
API id moonshotai.kimi-k3.1
Input / 1M
$4.40
at Amazon Bedrock
Output / 1M
$22.00
at Amazon Bedrock
Context window
869K
tokens
Max output
131K
tokens per reply
Hosts
1
4 listings
Released
-
date not published

Pricing

List prices at Amazon Bedrock, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$4.40$22.00
Cached input
95% off input
$0.22
Cache write, 5 minutes$5.50
Flex
50% off
$2.20$11.00
Priority
75% more
$7.70$38.50

What it costs a month

At Amazon Bedrock list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$2,594
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$1,901
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$3,089
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$15,840
Use your own numbers

Where to run it

1 host serve Kimi K3.1. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
moonshotai.kimi-k3.1
$4.40 / $22.00
Show 3 more listings: regions, snapshots and variants
Amazon Bedrock
global
global.moonshotai.kimi-k3.1
$4.00 / $20.00
−9% vs headline
in
in.moonshotai.kimi-k3.1
$4.40 / $22.00
us
us.moonshotai.kimi-k3.1
$4.40 / $22.00

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
TextImages
Produces
Text
Effort settings the source lists
nonelowmediumhighxhighmax
Features
Prompt caching
Open weights
License: Modified MIT License
Runtimes: SGLang, vLLM

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

  • Amazon Bedrock
    global
    $4.00 / $20.00
    9% less

Smaller Moonshot AI models

No cheaper Moonshot AI model keeps its key capabilities.

Other frontier labs

  • Claude Opus 5.5
    Anthropic · on Anthropic
    $4.00 / $20.00
    Has every key capability
    9% less
  • GPT-5.3 Codex
    OpenAI · on OpenAI
    $1.75 / $14.00
    Has every key capability
    45% less
  • Gemini 4 Carbon
    Google · on Google AI Studio
    $2.00 / $10.00
    Has every key capability
    55% less
  • Grok 4.7
    xAI · on xAI
    $2.00 / $6.00
    Has every key capability
    66% less

Open-weight picks

  • Kimi K3.1 Pro
    Moonshot AI · on Databricks
    $2.00 / $10.00
    Has every key capability
    55% less
  • Qwen4 Max
    Alibaba Qwen · on Together AI
    $1.43 / $4.28
    Has every key capability
    76% less
  • Inkling
    Independent · on DeepInfra
    $0.95 / $4.05
    Has every key capability
    80% less
  • DeepSeek Flash
    DeepSeek · on DeepSeek
    $0.30 / $1.20
    Has every key capability
    94% less

Model line

Newer and older versions, at each lab's own price.

  1. UndatedKimi K3.1Latest$4.40 / $22.00
  2. Jul 2026Kimi K3$3.00 / $15.00
  3. UndatedKimi K2.7$0.95 / $4.00
  4. Apr 2026Kimi K2.6$0.95 / $4.00
  5. Jan 2026Kimi K2.5$0.60 / $3.00
  6. Jul 2025Kimi K2$0.55 / $2.20

Call it

Using Amazon Bedrock's API.

Python
from litellm import completion

response = completion(
    model="bedrock/moonshotai.kimi-k3.1",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does Kimi K3.1 cost?

On Amazon Bedrock, Kimi K3.1 costs $4.40 per million input tokens and $22.00 per million output tokens. Cached input costs $0.22 per million tokens. The cheapest host is Amazon Bedrock (global), at $4.00 per million input tokens and $20.00 per million output tokens.

Which APIs offer Kimi K3.1?

Kimi K3.1 runs on 1 API: Amazon Bedrock.

What is Kimi K3.1's context window?

869K tokens, with up to 131K tokens in one reply.

Is Kimi K3.1 being retired?

No retirement date is scheduled for it on any API we track.

Is there a newer version of Kimi K3.1?

No. Kimi K3.1 is the newest model in its line.

What are cheaper alternatives to Kimi K3.1?

Models that cost less and keep most of what it can do: Claude Opus 5.5 on Anthropic (9% less), GPT-5.3 Codex on OpenAI (45% less) and Kimi K3.1 Pro on Databricks (55% less). Kimi K3.1 itself costs 9% less on Amazon Bedrock (global). Savings are on a mix of 3 input tokens to 1 output token.

What is Kimi K3.1's API model id?

Use moonshotai.kimi-k3.1 with Amazon Bedrock.

Sources

Raw catalog entry for moonshotai.kimi-k3.1 on Amazon Bedrock
{
  "cache_creation_input_token_cost": 0.0000055,
  "cache_creation_input_token_cost_flex": 0.00000275,
  "cache_creation_input_token_cost_priority": 0.000009625,
  "cache_read_input_token_cost": 2.2e-7,
  "cache_read_input_token_cost_flex": 1.1e-7,
  "cache_read_input_token_cost_priority": 3.85e-7,
  "input_cost_per_token": 0.0000044,
  "input_cost_per_token_flex": 0.0000022,
  "input_cost_per_token_priority": 0.0000077,
  "litellm_provider": "bedrock_converse",
  "max_input_tokens": 868928,
  "max_output_tokens": 131072,
  "mode": "chat",
  "output_cost_per_token": 0.000022,
  "output_cost_per_token_flex": 0.000011,
  "output_cost_per_token_priority": 0.0000385,
  "reasoning_effort_levels": [
    "none",
    "low",
    "medium",
    "high",
    "xhigh",
    "max"
  ],
  "source": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-moonshot-ai-kimi-k3-1-highspeed.md",
  "supports_audio_input": false,
  "supports_bedrock_converse_cache_point": false,
  "supports_function_calling": true,
  "supports_none_reasoning_effort": true,
  "supports_prompt_cache_breakpoint": false,
  "supports_prompt_caching": true,
  "supports_reasoning": true,
  "supports_regex_lookaround": false,
  "supports_response_schema": true,
  "supports_system_messages": true,
  "supports_tool_choice": true,
  "supports_vision": true
}