Kimi K3 (Low) is among the least intelligent models and particularly expensive when comparing to other open weight models of similar size. It's also slower than average per task, however highly concise. The model supports text and image input, outputs text, and has a 1M tokens context window.

Kimi K3 (Low) scores 17 on the Artificial Analysis Intelligence Index, placing it at the lower end among comparable models (median: 23). When evaluating the Intelligence Index, it generated 32.1M tokens, which is very concise in comparison to the median of 99.3M.

Pricing for Kimi K3 (Low) is $3.00 per 1M input tokens (expensive, median: $0.43) and $15.00 per 1M output tokens (expensive, median: $1.84). On average, it costs $2.37 per task to evaluate Kimi K3 (Low) on the Intelligence Index.

At 16.9 minutes per Intelligence Index task, Kimi K3 (Low) is slower than average compared with the median of 13.1 minutes for comparable models (lower is better).

Reasoning	Yes This page shows the reasoning version of this model. A non-reasoning variant may also exist.
Input modality	Supports: text and image
Output modality	Supports: text
Context window	1M ~1573 A4 pages of size 12 Arial font
Total parameters	2.8T
Active parameters	104B Number of parameters active per token during inference
License	Kimi K3 License
Model weights	Hugging Face

Metrics are compared against models of the same class:

Non-reasoning models → compared only with other non-reasoning models
Reasoning models → compared across both reasoning and non-reasoning
Open weights models → compared only with other open weights models of the same size class:

Tiny: ≤4B parameters
Small: 4B–40B parameters
Medium: 40B–150B parameters
Large: >150B parameters

Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:

<$0.15 per 1M tokens
$0.15–$1 per 1M tokens
>$1 per 1M tokens

Updated

Intelligence

Artificial Analysis Intelligence Index · Higher is better

No publicly available API

Time per Task

Inference time (minutes) per Intelligence Index task · Lower is better

No publicly available API

Cost per Task

Weighted average cost (USD) per Intelligence Index task · Lower is better

No publicly available API

IntelligenceUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v5.0.1 incorporates 9 evaluations: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), and Terminal-Bench-Science 0.1

No publicly available API

Artificial Analysis Intelligence Index v5.0.1 includes: AA-Briefcase v1.1, AutomationBench-AA, Terminal-Bench 4.0, FrontierCode-AA 2.0, AA-Omniscience v1.1, GDP.pdf, HLE-Diamond, CritPt v2 (Hard Subset), Terminal-Bench-Science 0.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

No publicly available API

Indicates whether the model weights are available. Models are labelled as 'Commercial Use Restricted' if commercial use is limited by conditions, and as 'Non-commercial' if the license prohibits commercial use.

Intelligence Index vs. Cost per Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task

Most attractive quadrant

Pareto line

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

Capability Indexes

Measures the performance of models on specific capabilities and industries

Artificial Analysis Finance & Accounting Index

Incorporates 5 evaluations: AA-Briefcase v1.1, AA-Omniscience v1.1, HLE-Diamond, AutomationBench-AA, GDP.pdf · Higher is better

No publicly available API

Benchmarks

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100

No publicly available API

AA-Briefcase v1.1

Agentic knowledge work, Elo

AA-Omniscience v1.1 AccuracyUpdated

Knowledge

AA-Omniscience v1.1 Non-Hallucination RateUpdated

1 - hallucination rate

AutomationBench-AA

Agentic SaaS workflows

Terminal-Bench-Science 0.1New

Agentic scientific research workflows in a terminal

FrontierCode-AA 2.0New

Agentic software engineering in open-source repositories

HLE-DiamondNew

Reasoning & knowledge

GDP.pdf

Professional document reasoning, All-pass

Terminal-Bench 4.0

Agentic coding & terminal use

CritPt v2 (Hard Subset)

Physics reasoning

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

Harvey LAB-AA v1.1Updated

Legal agentic work, Hallucination-Gated All-Pass Rate

EnterpriseOps-Gym-AA

Agentic business operations

GDPval-AA v2.1

Agentic real-world work tasks, Elo

AA-3DNew

Agentic interactive 3D experiences, Elo

Code2Video-AANew

Product and brand videos designed in code, Elo

ITBench-AA

Kubernetes incident root-cause analysis

MLCR-AAUpdated

Medical long context reasoning

MMMU-Pro

Visual reasoning

AA-PersonalityBench: Expressiveness ScoreNew

Rich, varied, well-crafted language

AA-PersonalityBench: Organization ScoreNew

Well-structured, scannable, answer-first

AA-PersonalityBench: Transparency ScoreNew

Truthful, acknowledges uncertainty, non-sycophantic

AA-PersonalityBench: Slop ScoreNew

Use of filler and generic language

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

AA-Briefcase v1.1

AA-Briefcase Elo

AA-Briefcase v1.1 is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better

No publicly available API

AA-Briefcase Elo is a combined metric that aggregates analytical quality Elo, presentation Elo, and rubric pass rate, with rubric performance converted into Elo via synthetic head-to-head matches. Elo and 95% confidence interval bounds are clamped at 0.

AA-Omniscience v1.1

AA-Omniscience Net Score

AA-Omniscience v1.1 Net Score (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct.

No publicly available API

Openness Index

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open)

Token Use

Output Tokens per Intelligence Index Task

Weighted average number of output tokens used to run one task in the Artificial Analysis Intelligence Index

No publicly available API

The number of tokens required per Intelligence Index task. This is calculated by multiplying the output tokens per eval by the relative weights of each benchmark in the Intelligence Index, then dividing by task count (excluding repeats).

Cost

Cost per Intelligence Index Task

Weighted average cost (USD) per Intelligence Index task · Segmented by token type · Lower is better

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index

The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input, cache hit, cache write, reasoning, and answer token prices and the number of tokens used across evaluations (excluding repeats).

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens)

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Context Window

Context window: tokens limit · Higher is better

No publicly available API

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Performance

Measured by time per task (minutes), output speed (tokens per second), time to first answer token and time to first token (seconds)

Time per Intelligence Index Task

Output tokens divided by end-to-end output speed (minutes per Intelligence Index task) · Speed at the model's own endpoint or the median across providers · Lower is better

The weighted average time (minutes) per Artificial Analysis Intelligence Index task: output tokens per task divided by output speed, weighted by each benchmark's weight in the Intelligence Index.

Speed and latency metrics average each endpoint's runs over the past 5 days, and the error rate pools them. We run the benchmark every 6 hours, and endpoints of deprecated models every 18 hours.

Output Speed

Output tokens divided by total request time, including time to first token · Higher is better

Output tokens, including reasoning tokens, received per second over a request's full duration, from sending the request to receiving the final token. The headline value averages each Performance v3 workload over the past 5 days, then averages the workloads.

Time to First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time

Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.

Model Size (Open Weights Models Only)

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference (billions)

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

The number of parameters actually executed during each inference forward pass, expressed in billions. For Mixture of Experts (MoE) models, a routing mechanism selects a subset of experts per token, resulting in fewer active than total parameters. Dense models use all parameters, so active equals total.

Frequently Asked Questions

Common questions about Kimi K3 (Low)

Kimi K3 (Low) was released on July 16, 2026.

Kimi K3 (Low) was created by Kimi.

Kimi K3 (Low) scores 17 on the Artificial Analysis Intelligence Index, placing it below average among other open weight models of similar size (median: 23).

Kimi K3 (Low) takes 16.9 minutes per Intelligence Index task (based on Kimi's API), which is slower than average compared to other open weight models of similar size (median: 13.1 minutes; lower is better).

Kimi K3 (Low) has a time to first answer token of 10.60s (based on Kimi's API), which is better than average compared to other open weight models of similar size (median: 20.03s).

Kimi K3 (Low) costs $3.00 per 1M input tokens (at the higher end, median: $0.43) and $15.00 per 1M output tokens (at the higher end, median: $1.84), based on Kimi's API.

Kimi K3 (Low) costs $3.00 per 1M input tokens and $15.00 per 1M output tokens (based on Kimi's API). For a blended rate (7:2:1 cache hit/input/output ratio), this is $2.31 per 1M tokens. Pricing may vary by provider. Compare provider pricing

When evaluated on the Intelligence Index, Kimi K3 (Low) generated 32.1M output tokens, which is very competitive compared to other open weight models of similar size (median: 99.3M).

Yes, Kimi K3 (Low) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.

Kimi K3 (Low) supports text and image input.

Kimi K3 (Low) supports text output.

Yes, Kimi K3 (Low) supports image input and can analyze, describe, and answer questions about images.

Yes, Kimi K3 (Low) is multimodal. It can process text and image input and generate text output.

Kimi K3 (Low) has a context window of 1M tokens. This determines how much text and conversation history the model can process in a single request.

Yes, Kimi K3 (Low) is open weights. The model weights are publicly available and can be downloaded for self-hosting.

Kimi K3 (Low) has 2.8T parameters (104B active).

Kimi K3 (Low) is a Mixture of Experts (MoE) model with 2.8T total parameters, but only 104B active parameters are used during inference.

Kimi K3 (Low) is released under the Kimi K3 License license. Commercial use is allowed with restrictions. View license

Kimi K3 (Low) achieves a score of 17 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across knowledge work, coding, science, and general knowledge.

Yes, Kimi K3 (Low) is available via API through 6 providers. Compare API providers

Kimi K3 (Low) is available through 6 API providers. Compare providers

Kimi K3 (Low) Intelligence, Performance & Price Analysis

Model summary

IntelligenceUpdated

Comparison Summary

Technical specifications

129 models in this class

IntelligenceUpdated

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index

Open Weights

Intelligence Index vs. Cost per Task

Cost per Intelligence Index Task

Artificial Analysis Intelligence Index

Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

AA-Briefcase v1.1

AA-Briefcase Elo

AA-Briefcase Elo

AA-Omniscience v1.1

AA-Omniscience Net Score

AA-Omniscience Net Score

Openness Index

Artificial Analysis Openness Index: Score

Token Use

Output Tokens per Intelligence Index Task

Output Tokens per Intelligence Index Task

Cost

Cost per Intelligence Index Task

Cost per Intelligence Index Task

Cost to Run Artificial Analysis Intelligence Index

Cost to Run Artificial Analysis Intelligence Index

Pricing: Cache Hit, Input, and Output

Cache Hit

Context Window

Context Window

Context Window for RAG

Context Window

Performance

Time per Intelligence Index Task

Time per Intelligence Index Task

Time Period

Output Speed

End-to-End Output Speed

Time to First Answer Token

Time to First Answer Token

Model Size (Open Weights Models Only)

Model Size: Total and Active Parameters

Total Parameters

Active Parameters at Inference Time

Frequently Asked Questions

When was Kimi K3 (Low) released?

Who created Kimi K3 (Low)?

How intelligent is Kimi K3 (Low)?

How fast is Kimi K3 (Low)?

What is the latency of Kimi K3 (Low)?

How much does Kimi K3 (Low) cost?

What is Kimi K3 (Low) API pricing?

How verbose is Kimi K3 (Low)?

Is Kimi K3 (Low) a reasoning model?

What input modalities does Kimi K3 (Low) support?

What output modalities does Kimi K3 (Low) support?

Can Kimi K3 (Low) process images?

Is Kimi K3 (Low) multimodal?

What is the context window of Kimi K3 (Low)?

Is Kimi K3 (Low) open source?

How many parameters does Kimi K3 (Low) have?

What are the active parameters of Kimi K3 (Low)?

What is the license for Kimi K3 (Low)?

How does Kimi K3 (Low) perform on benchmarks?

Is Kimi K3 (Low) available via API?

Where can I use Kimi K3 (Low)?