ERNIE Bot
BaiduChina
25
Baidu's flagship assistant, now on open-weight ERNIE models.
- Version
- 5.0-thinking-preview
- Cost
- Pay-as-you-go API pricing based on tokens. Token Plan Personal Edition and some free trials available.
- Model
- DeepSeek-V3.1
Kimi
Moonshot AI · #4 most active of 27 in Writing
Moonshot AI's assistant, known for very long context windows and the open-weight K2 agentic model.
100
Trust score · Verified
3 data points checked · 2026-09-12
Current version
Kimi K3
Entry cost
API access with pay-as-you-go pricing based on token consumption. File APIs are currently free.
Changes / 30d
10
The short answers · verified September 16, 2026
- What is the latest version of Kimi?
- The current shipped version of Kimi is Kimi K3, as of September 16, 2026.
- How much does Kimi cost?
- Kimi has no published flat entry price. API access with pay-as-you-go pricing based on token consumption. File APIs are currently free.
- What AI model does Kimi use?
- Kimi runs primarily on Kimi K2.6.
Source: Moonshot AI — platform.moonshot.cn · Reusable under CC BY 4.0 — cite Tomorrow
Capabilities
- Native multimodal (visual + text)
- 1M-token context
- Agentic tool use
- Thinking / non-thinking modes
- Dedicated coding models (k2.7-code, high-speed)
- Open weights
- 1M token context
- Multimodal input
- Vision understanding
- JSON Mode
- Tool calling
- Web search
- Reasoning / Thinking mode
- Streaming output
- Context caching
- Batch API
- multimodal input
- streaming output
- reasoning model
- tool calling
- json mode
- context caching
- web search
- 1M Token Context
- Multimodal Input (Image/Video)
- Reasoning Mode
- Tool Calling
- Context Caching
- OpenAI Compatibility
- Reasoning Model
- Multimodal Input
- Web Search
- Streaming
- File Q&A
- Thinking Mode
- Search Integration
- Text Input
- Image InputExternal
- Video Input
- visual understanding
- reasoning mode
- context cache
- batch api
- video input
- Vision Understanding
- Stream Output
- Thinking Model
- File QA
- Visual Input (Image/Video)
- Multi-turn Dialogue
- Streaming Output
- Multi-modal input
- Function calling
- Thinking/reasoning mode
- Vision/Video Input
- Text input
- Visual input (Image/Video)
- Web search integration
- Text Generation
- Visual Understanding
- Image input
- Video input
- Thinking mode
- Structured Outputs
- API
- Multimodal
- Function Calling
- Reasoning
- Context Cache
- Image Input
- Visual understanding
- Vision Input
- Multi-turn conversation
- Visual input
- Multi-turn chat
- vision understanding
- thinking model
- multi-turn conversation
- Multi-turn Conversation
- Partial Mode
- Visual Input
- text input
- image input
- streaming
- reasoning
- openai-compatible api
- stream output
- 视觉输入
- 上下文缓存
- 工具调用
- 思考模型
- 联网搜索
- 流式输出
- tools
- thinking mode
- vision_understanding
- context_caching
- tool_calling
- web_search
- json_mode
- batch_api
- thinking_mode
- Code Generation
Pricing
Pay-as-you-go / Pay-per-token
Charged per million tokens, file upload and parsing interfaces are temporarily free.
Free
Entry price over time
Geek mode
Models under the hood
- disclosed
Kimi K2.6
Moonshot AI
- disclosed
Kimi K2.7 Code
Moonshot AI
- disclosed
Kimi K3
Moonshot AI
- disclosed
kimi-k2.5
Moonshot AI
- disclosed
kimi-k2.7-code-highspeed
Moonshot AI
Context window
1000000
Public API
yes
Multi-model routing
yes
Stack signals
- openai
Shared model stack
Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.
- Kimi-K3+1 aliasMoonshot AI · 4 products
- Kimi K2.5+1 aliasMoonshot AI · 3 products
- Kimi K2.6Moonshot · 3 products
- Kimi K2.7 CodeMoonshot AI · 2 products
- kimi-k2.7-code-highspeedMoonshot AI · 1 product
green disclosed · amber inferred · grey unattributed. Aliases are folded into one model; provider concentration counts only models with an established vendor.
Reported scale
Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.
- Third-party estimate
Monthly active users
36M
monthly users
as of Mar 2025·QuestMobile China app report ·inferred
Also in Writing
All WritingHunyuan
TencentChina
16
Tencent's multimodal model family and assistant.
- Version
- Hunyuan-role-latest
- Cost
- Free trial up to 1 million tokens. Pay-as-you-go pricing applies afterward.
- Model
- Hunyuan-a13b
ChatGPT
OpenAI
12
General-purpose assistant
- Version
- GPT-6 Astra
- Cost
- Pricing not published in the provided documentation
- Model
- GPT Image
Notion AI
Notion
10
AI inside your workspace
- Version
- 3.7
- Cost
- Free tier · from $10/mo
- Model
- Claude Sonnet 4.5
Qwen Chat
AlibabaChina
10
Alibaba's assistant on the widely forked Qwen model family.
- Version
- Qwen3-2507
- Cost
- Free to chat on official website
- Model
- Qwen3
DeepSeek
DeepSeekChina
9
Open-weight reasoning models at a fraction of frontier pricing.
- Version
- DeepSeek-V4.1-Flash
- Cost
- Pay-as-you-go API pricing based on token consumption. No flat monthly fee.
- Model
- DeepSeek-R1
Change history
- capability
- capability
- capability
New capabilities: vision_understanding, context_caching, tool_calling
95 tracked102 tracked · +vision_understanding, context_caching, tool_calling, web_search, json_mode, batch_api, thinking_mode
source - capability
- capability
- capability
- capability
- capability
- capability
New capabilities: 视觉输入, 上下文缓存, 工具调用
87 tracked93 tracked · +视觉输入, 上下文缓存, 工具调用, 思考模型, 联网搜索, 流式输出
source - capability
- capability
- capability
- capability
- capability
- capability
New capabilities: text input, image input, streaming
81 tracked86 tracked · +text input, image input, streaming, reasoning, openai-compatible api
source - capability
- capability
New capabilities: Multi-turn Conversation, Partial Mode, Visual Input
78 tracked81 tracked · +Multi-turn Conversation, Partial Mode, Visual Input
source - capability
New capabilities: vision understanding, thinking model, multi-turn conversation
75 tracked78 tracked · +vision understanding, thinking model, multi-turn conversation
source - capability
- capability
- capability
- capability
New capabilities: Multi-turn conversation, Visual input
72 tracked74 tracked · +Multi-turn conversation, Visual input
source - capability
- capability
- capability
- capability
- capability
- capability
- capability
- capability
New capabilities: API, Multimodal, Function Calling
64 tracked68 tracked · +API, Multimodal, Function Calling, Reasoning
source - capability
- capability
- capability
- capability
New capabilities: Image input, Video input, Thinking mode
60 tracked63 tracked · +Image input, Video input, Thinking mode
source - capability
- capability
New capabilities: Text Generation, Visual Understanding
58 tracked60 tracked · +Text Generation, Visual Understanding
source - capability
- capability
New capabilities: Text input, Visual input (Image/Video), Web search integration
55 tracked58 tracked · +Text input, Visual input (Image/Video), Web search integration
source - capability
- capability
Kimi compared
Straight head-to-head pages against the busiest products in Writing.
How to cite this page
Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/tools/kimi
APA
Tomorrow. (2026). Kimi — version, pricing and model stack [Data set entry]. AlienSquad. Retrieved 2026-09-17, from https://tomorrow.aliensquad.ai/tools/kimi
BibTeX
@misc{tomorrow-tools-kimi,
author = {{Tomorrow}},
title = {Kimi — version, pricing and model stack},
year = {2026},
publisher = {AlienSquad},
howpublished = {\url{https://tomorrow.aliensquad.ai/tools/kimi}},
note = {Accessed: 2026-09-17}
}