How much does Kimi K3.1 Pro cost?

Kimi K3.1 Pro costs $0.2375/M input tokens and $8.411/M output tokens, with separate rates for Cache Read at $0.20/M tokens.

What is the context length of Kimi K3.1 Pro?

Kimi K3.1 Pro has a 1,048,576 token context window. It supports up to 1,048,576 completion tokens.

Does Kimi K3.1 Pro support tool calling and structured outputs?

Yes. Kimi K3.1 Pro accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.

What inputs and outputs does Kimi K3.1 Pro support?

Kimi K3.1 Pro accepts text, images, and video as input and returns text.

Which providers serve Kimi K3.1 Pro?

Kimi K3.1 Pro is served by 18 providers on OpenRouter: Relace, Wafer, Morph, Sail Research, AkashML, Decart, inference.net, DeepInfra and 10 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.

When was Kimi K3.1 Pro released?

Kimi K3.1 Pro was released on October 23, 2026.

MoonshotAI: Kimi K3.1 Pro

moonshotai/kimi-k3.1-pro

Model weights

Compare

Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window. It is built for long-horizon coding, 3D creation, financial research, and end-to-end agentic knowledge work, with native image and video understanding.

Modalities

In / out price

$0.2375 / $8.411per 1M

Context

1.0M

Released

Oct 23, 2026

Providers

Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).

Latency / throughput


	$0.253	$8.80	$0.18	0.42s	96 tok/s	99.97%
	$2.00	$10.00	$0.20	0.72s	95 tok/s	100.00%
	$2.00	$10.00	$0.20	1.36s	108 tok/s	99.95%
	$1.80	$9.00	$0.18	0.77s	103 tok/s	100.00%
	$1.80	$9.00	$0.18	0.75s	83 tok/s	100.00%
	$0.2375	$8.4114	$0.20	1.03s	61 tok/s	100.00%
	$2.00	$10.00	$0.20	1.08s	75 tok/s	99.63%
Fast	$2.70	$13.50	$0.27	0.75s	118 tok/s	100.00%
Fast	$2.70	$13.50	$0.27	0.87s	118 tok/s	100.00%
	$2.00	$10.00	$0.20	1.17s	107 tok/s	93.15%
	$0.871	$11.111	$0.278	1.60s	68 tok/s	99.72%
12% off	$2.00$1.76	$10.00$8.80	$0.20$0.176	1.33s	63 tok/s	99.87%
	$2.00	$10.00	$0.20	0.55s	70 tok/s	94.39%
	$2.70	$13.50	$0.27	1.25s	98 tok/s	99.39%
	$2.00	$10.00	$0.20	1.46s	56 tok/s	98.96%
Highspeed	$4.00	$20.00	$0.20	2.99s	58 tok/s	100.00%
	$2.00	$10.00	$0.20	2.38s	31 tok/s	100.00%
	$2.00	$10.00	$0.20	1.48s	41 tok/s	93.10%
	$1.80	$9.00	$0.18	0.98s	91 tok/s	100.00%
	$2.30	$11.50	$0.23	2.05s	39 tok/s	98.46%
	$1.0425	$9.35	$0.187	2.64s	42 tok/s	99.90%
	$1.4025	$9.35	$0.187	3.93s	47 tok/s	99.95%
	$1.75	$10.00	$0.18	3.26s	38 tok/s	99.70%
5% off	$2.00$1.90	$10.00$9.50	$0.20$0.19	1.11s	45 tok/s	99.84%

Input $0.253Output $8.80
96 tok/s99.97% uptime
Input $2.00Output $10.00
95 tok/s100.00% uptime
Input $2.00Output $10.00
108 tok/s99.95% uptime
Input $1.80Output $9.00
103 tok/s100.00% uptime
Input $1.80Output $9.00
83 tok/s100.00% uptime
Input $0.2375Output $8.4114
61 tok/s100.00% uptime
Input $2.00Output $10.00
75 tok/s99.63% uptime
Input $2.70Output $13.50
Fast
118 tok/s100.00% uptime
Input $2.70Output $13.50
Fast
118 tok/s100.00% uptime
Input $2.00Output $10.00
107 tok/s93.15% uptime
Input $0.871Output $11.111
68 tok/s99.72% uptime
Input $1.76Output $8.80
12% off
63 tok/s99.87% uptime
Input $2.00Output $10.00
70 tok/s94.39% uptime
Input $2.70Output $13.50
98 tok/s99.39% uptime
Input $2.00Output $10.00
56 tok/s98.96% uptime
Input $4.00Output $20.00
Highspeed
58 tok/s100.00% uptime
Input $2.00Output $10.00
31 tok/s100.00% uptime
Input $2.00Output $10.00
41 tok/s93.10% uptime
Input $1.80Output $9.00
91 tok/s100.00% uptime
Input $2.30Output $11.50
39 tok/s98.46% uptime
Input $1.0425Output $9.35
42 tok/s99.90% uptime
Input $1.4025Output $9.35
47 tok/s99.95% uptime
Input $1.75Output $10.00
38 tok/s99.70% uptime
Input $1.90Output $9.50
5% off
45 tok/s99.84% uptime

Pricing

The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.

Weighted average

Input

$0.3678/M tokens

Output

$10.23/M tokens

Kimi K3.1 Pro — Price history

Price history

No effective price data for this selection

	$0.191	$9.00	84.8%	14.6%
	$0.203	$8.41	90.8%	7.6%
	$0.255	$9.35	92.0%	0.8%
	$0.285	$9.00	93.5%	2.9%
	$0.292	$9.00	93.1%	5.1%
	$0.299	$9.35	90.8%	0.1%
	$0.335	$10.00	92.5%	9.3%
	$0.338	$11.15	90.9%	4.5%
	$0.345	$10.00	92.0%	8.9%
	$0.358	$8.80	88.5%	1.4%
	$0.362	$10.00	91.0%	0.6%
	$0.388	$10.00	89.6%	5.0%
	$0.389	$9.00	87.1%	3.0%
	$0.407	$10.00	88.5%	5.3%
Fast	$0.408	$13.50	94.3%	4.8%
Fast	$0.426	$13.50	93.6%	3.5%
Highspeed	$0.452	$20.00	93.4%	1.8%
	$0.497	$10.00	83.5%	0.6%
	$0.499	$10.00	83.4%	2.8%
	$0.516	$10.00	78.6%	0.8%
	$0.520	$11.16	86.2%	0.9%
	$0.548	$9.52	79.2%	0.2%
	$0.601	$10.00	77.7%	0.5%
	$0.602	$11.50	82.0%	14.8%

Effective in $0.191Effective out $9.00
Token share
14.6%
Effective in $0.203Effective out $8.41
Token share
7.6%
Effective in $0.255Effective out $9.35
Token share
0.8%
Effective in $0.285Effective out $9.00
Token share
2.9%
Effective in $0.292Effective out $9.00
Token share
5.1%
Effective in $0.299Effective out $9.35
Token share
0.1%
Effective in $0.335Effective out $10.00
Token share
9.3%
Effective in $0.338Effective out $11.15
Token share
4.5%
Effective in $0.345Effective out $10.00
Token share
8.9%
Effective in $0.358Effective out $8.80
Token share
1.4%
Effective in $0.362Effective out $10.00
Token share
0.6%
Effective in $0.388Effective out $10.00
Token share
5.0%
Effective in $0.389Effective out $9.00
Token share
3.0%
Effective in $0.407Effective out $10.00
Token share
5.3%
Effective in $0.408Effective out $13.50
FastToken share
4.8%
Effective in $0.426Effective out $13.50
FastToken share
3.5%
Effective in $0.452Effective out $20.00
HighspeedToken share
1.8%
Effective in $0.497Effective out $10.00
Token share
0.6%
Effective in $0.499Effective out $10.00
Token share
2.8%
Effective in $0.516Effective out $10.00
Token share
0.8%
Effective in $0.520Effective out $11.16
Token share
0.9%
Effective in $0.548Effective out $9.52
Token share
0.2%
Effective in $0.601Effective out $10.00
Token share
0.5%
Effective in $0.602Effective out $11.50
Token share
14.8%

Performance

Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).

Throughput

118tok/s

P50, best across providers

Latency

0.42s

P50, best provider

Reasoning effortLatency / throughput

AutoExacto benchmarks

GPQA DiamondTAU-BenchDigitalOcean93.3%78.7%Moonshot AI (Highspeed)93.3%78.7%Phala92.9%78.7%Fireworks92.8%78.7%Nebius Token Factory93.9%77.3%Moonshot AI92.6%75.3%Phala92.4%75.0%

+14 more providers

Uptime

Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.

Uptime (3d)

100.00%

Availability (3d)

99.90%

Availability over the last 3 days

Last 72 hours

Availability 99.90%

3 Days Ago2 Days AgoYesterdayNow

Availability over the last 24 hours

OpenRouter Availability

99.90%

Without Routing

95.27%

When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.

Benchmarks

Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.

Benchmark score summary for MoonshotAI: Kimi K3.1 Pro (Artificial Analysis, Design Arena and Vals AI)
Source	Benchmark	Score
Artificial Analysis	Kimi K3.1 Pro Intelligence Index	46.5
Artificial Analysis	Kimi K3.1 Pro HLE	52.5%
Artificial Analysis	Kimi K3.1 Pro AA-LCR	86.3%
Artificial Analysis	Kimi K3.1 Pro GDPval-AA	68.0%
Artificial Analysis	Kimi K3.1 Pro CritPt	24.6%
Artificial Analysis	Kimi K3.1 Pro SciCode	60.8%
Artificial Analysis	Kimi K3.1 Pro Terminal-Bench 4.0	67.2%
Artificial Analysis	Kimi K3.1 Pro AA-Omniscience Accuracy	48.9%
Artificial Analysis	Kimi K3.1 Pro AA-Omniscience Non-Hallucination Rate	62.7%
Design Arena	Kimi K3.1 Pro Agents Arena Androidnative Elo	1149
Design Arena	Kimi K3.1 Pro Agents Arena Codetovideo Elo	1233
Design Arena	Kimi K3.1 Pro Agents Arena Full Stack Elo	1326
Design Arena	Kimi K3.1 Pro Agents Arena Mobile Apps Elo	1272
Design Arena	Kimi K3.1 Pro Agents Arena Pdf Elo	1229
Design Arena	Kimi K3.1 Pro Agents Arena Python-Pptxslides Elo	1292
Design Arena	Kimi K3.1 Pro Agents Arena Webapps Elo	1303
Design Arena	Kimi K3.1 Pro Models Arena 3D Elo	1440
Design Arena	Kimi K3.1 Pro Models Arena Asciiart Elo	1301
Design Arena	Kimi K3.1 Pro Models Arena Code Categories Elo	1350
Design Arena	Kimi K3.1 Pro Models Arena Data Visualization Elo	1298
Design Arena	Kimi K3.1 Pro Models Arena Game Development Elo	1404
Design Arena	Kimi K3.1 Pro Models Arena SVG Elo	1330
Design Arena	Kimi K3.1 Pro Models Arena UI Component Elo	1391
Design Arena	Kimi K3.1 Pro Models Arena Website Elo	1321
Vals AI	Kimi K3.1 Pro Access to Justice Benchmark	69.5%
Vals AI	Kimi K3.1 Pro BioMysteryBench	75.6%
Vals AI	Kimi K3.1 Pro CyberOffense	53.8%
Vals AI	Kimi K3.1 Pro EMB v1.1	70.0%
Vals AI	Kimi K3.1 Pro Finance Agent v2.1	61.5%
Vals AI	Kimi K3.1 Pro Vals Finance Index	56.0%
Vals AI	Kimi K3.1 Pro HLAB-Vals-1.1	11.7%
Vals AI	Kimi K3.1 Pro IOI	81.9%
Vals AI	Kimi K3.1 Pro Vals Legal Index	67.0%
Vals AI	Kimi K3.1 Pro Legal Research Bench	37.5%
Vals AI	Kimi K3.1 Pro MysteryMechanism	26.1%
Vals AI	Kimi K3.1 Pro ProofBench v1.1	91.0%
Vals AI	Kimi K3.1 Pro Vals Science Index	44.4%
Vals AI	Kimi K3.1 Pro SRE Bench	27.1%
Vals AI	Kimi K3.1 Pro Tax Agent Bench	24.9%
Vals AI	Kimi K3.1 Pro Terminal-Bench 4.0	63.6%
Vals AI	Kimi K3.1 Pro Terminal-Bench-Science	31.4%
Vals AI	Kimi K3.1 Pro Vals Index	62.5%
Vals AI	Kimi K3.1 Pro Vibe Code Bench v1.1	84.9%

Apps

Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.

There are many coding agents, but this one is yours.

230Btokens

Hermes Agent

Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.

208Btokens

Claude Code

Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.

185Btokens

omp

new

133Btokens

Codex

A coding agent that helps you build and ship with AI

124Btokens

Rest of OpenRouter

341B

Top 5 Apps

41B

Not currently available through EU in-region routing. In-region routing docs

More models from moonshotai

Kimi K3.1 Pro (batch)

Text1.0M context$0.70 / $6

Kimi K3

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.

Text1.0M context$0.1824 / $11.01

Kimi K3 (batch)

Text1.0M context$2.28 / $11.40

Kimi K2.7 Code

MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts. It uses a native multimodal mixture-of-experts architecture that accepts text and image input, and it always operates in a thinking mode, preserving full reasoning content across multi-turn conversations. With a 256K-token context window, it targets long-horizon coding, agentic task decomposition, and multi-turn dialogue. The model activates 32B parameters out of roughly 1T total.

Text262K context$0.60 / $3

Kimi Latest

This model always redirects to the latest model in the Kimi family.

Text262K context

Kimi K2.6

Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration. It handles complex end-to-end coding tasks across Python, Rust, and Go, and can convert prompts and visual inputs into production-ready interfaces. Its agent swarm architecture scales to hundreds of parallel sub-agents for autonomous task decomposition - delivering documents, websites, and spreadsheets in a single run without human oversight.

Text262K context$0.3895 / $1.64

Kimi K2.5

Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed visual and text tokens, it delivers strong performance in general reasoning, visual coding, and agentic tool-calling.

Text262K context$0.49 / $2.50

Kimi K2 Thinking

Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in Kimi K2, it activates 32 billion parameters per forward pass and supports 256 k-token context windows. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift.

It sets new open-source benchmarks on HLE, BrowseComp, SWE-Multilingual, and LiveCodeBench, while maintaining stable multi-agent behavior through 200–300 tool calls. Built on a large-scale MoE architecture with MuonClip optimization, it combines strong reasoning depth with high inference efficiency for demanding agentic and analytical tasks.

Text262K context$0.60 / $2.50

Kimi K2 0905

Kimi K2 0905 is the September update of Kimi K2 0711. It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports long-context inference up to 256k tokens, extended from the previous 128k.

This update improves agentic coding with higher accuracy and better generalization across scaffolds, and enhances frontend coding with more aesthetic and functional outputs for web, 3D, and related tasks. Kimi K2 is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. It excels across coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) benchmarks. The model is trained with a novel stack incorporating the MuonClip optimizer for stable large-scale MoE training.

Text262K context$0.60 / $2.50

Kimi K2 0711

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. Kimi K2 excels across a broad range of benchmarks, particularly in coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) tasks. It supports long-context inference up to 128K tokens and is designed with a novel training stack that includes the MuonClip optimizer for stable large-scale MoE training.

Text131K context$0.57 / $2.30

Kimi Dev 72B

Kimi-Dev-72B is an open-source large language model fine-tuned for software engineering and issue resolution tasks. Based on Qwen2.5-72B, it is optimized using large-scale reinforcement learning that applies code patches in real repositories and validates them via full test suite execution—rewarding only correct, robust completions. The model achieves 60.4% on SWE-bench Verified, setting a new benchmark among open-source models for software bug fixing and code reasoning.

Text131K context

Kimi VL A3B Thinking

Kimi-VL is a lightweight Mixture-of-Experts vision-language model that activates only 2.8B parameters per step while delivering strong performance on multimodal reasoning and long-context tasks. The Kimi-VL-A3B-Thinking variant, fine-tuned with chain-of-thought and reinforcement learning, excels in math and visual reasoning benchmarks like MathVision, MMMU, and MathVista, rivaling much larger models such as Qwen2.5-VL-7B and Gemma-3-12B. It supports 128K context and high-resolution input via its MoonViT encoder.

Text131K context

Moonlight 16B A3B Instruct

Moonlight-16B-A3B-Instruct is a 16B-parameter Mixture-of-Experts (MoE) language model developed by Moonshot AI. It is optimized for instruction-following tasks with 3B activated parameters per inference. The model advances the Pareto frontier in performance per FLOP across English, coding, math, and Chinese benchmarks. It outperforms comparable models like Llama3-3B and Deepseek-v2-Lite while maintaining efficient deployment capabilities through Hugging Face integration and compatibility with popular inference engines like vLLM12.

Text8K context

MoonshotAI: Kimi K3.1 Pro

Providers