Mini tools| Model | Input | Cached input | Output | Unit |
|---|---|---|---|---|
|
Deepseek V4 Pro
DeepSeek
|
$1.74 | $0.2 | $3.48 | per 1M tokens |
|
Gemma 4 31B IT NVFP4
NVIDIA
|
$0.28 | - | $0.86 | per 1M tokens |
|
GLM 5
Z.ai
|
$1 | - | $3.2 | per 1M tokens |
|
GLM 5.1
Z.ai
|
$1.4 | $0.26 | $4.4 | per 1M tokens |
|
Kimi K2.6
Moonshot AI
|
$1.2 | $0.2 | $4.5 | per 1M tokens |
|
Kimi K2.7 Code
This model
Moonshot AI
|
$0.95 | $0.19 | $4 | per 1M tokens |
|
LFM2 24B A2B
Liquid AI
|
$0.03 | - | $0.12 | per 1M tokens |
|
Llama 3.3
Meta Platforms
|
$1.04 | - | $1.04 | per 1M tokens |
|
MiniMax M2.5
MiniMax
|
$0.3 | $0.06 | $1.2 | per 1M tokens |
|
MiniMax M2.7
MiniMax
|
$0.3 | $0.06 | $1.2 | per 1M tokens |
|
MiniMax M3
MiniMax
|
$0.3 | $0.06 | $1.2 | per 1M tokens |
| $0.6 | $0.2 | $3.6 | per 1M tokens | |
|
Qwen 3.5 9B
Alibaba
|
$0.17 | - | $0.25 | per 1M tokens |
|
Qwen 3.6 Plus
Alibaba
|
$0.5 | - | $3 | per 1M tokens |
|
Qwen 3.7 Max
Alibaba
|
$1.25 | $0.13 | $3.75 | per 1M tokens |
|
Qwen3 235B A22B
Alibaba
|
$0.2 | - | $0.6 | per 1M tokens |
|
Qwen3.5 397B A17B
Alibaba
|
$0.6 | $0.35 | $3.6 | per 1M tokens |
|
Qwen3.7-Plus
Alibaba
|
$0.32 | - | $1.28 | per 1M tokens |
Kimi K2.7 Code
By Moonshot AI
Model family: Kimi
Kimi K2.7 Code is built on a Mixture-of-Experts architecture with 1 trillion total parameters and 32 billion active per token. It supports a 262,144-token context window with a native multimodal architecture accepting text, image, and video inputs (PNG, JPEG, WebP, GIF, MP4, MOV, AVI, and more). The model operates exclusively in thinking (reasoning) mode -- non-thinking mode is not supported. It excels at long-horizon coding tasks, complex logical reasoning, and mathematical problems, achieving a 21.8% improvement on Kimi Code Bench v2 over K2.6 while reducing overthinking by 30% and improving agentic performance by 10%. It supports multi-step tool calling, JSON Mode, Partial Mode, and automatic context caching. A high-speed variant (kimi-k2.7-code-highspeed) delivers approximately 180 tokens/second. Released under a Modified MIT license with weights on Hugging Face.
New
Multimodal
Overview
Kimi K2.7 Code is a coding-focused model built on a 1T parameter MoE architecture with 32B active parameters and a 256K context window. It features native multimodal input (text, image, video), mandatory long-thinking mode, tool calling, and enhanced agentic coding. It reduces overthinking by 30% vs K2.6 and achieves a 21.8% improvement on Kimi Code Bench v2.
Pricing
Compare Kimi K2.7 Code with other models listed in the same vendor pricing tiers and context lengths.
Tier
Standard
Batch
| Model | Input | Cached input | Output | Unit |
|---|---|---|---|---|
|
Deepseek V4 Pro
DeepSeek
|
$0.87 | $0.2 | $1.74 | per 1M tokens |
|
Gemma 4 31B IT NVFP4
NVIDIA
|
$0.28 | - | $0.86 | per 1M tokens |
|
GLM 5
Z.ai
|
$1 | - | $3.2 | per 1M tokens |
|
GLM 5.1
Z.ai
|
$0.7 | $0.26 | $2.2 | per 1M tokens |
|
Kimi K2.6
Moonshot AI
|
$1.2 | $0.2 | $4.5 | per 1M tokens |
|
Kimi K2.7 Code
This model
Moonshot AI
|
$0.95 | $0.19 | $4 | per 1M tokens |
|
LFM2 24B A2B
Liquid AI
|
$0.01 | - | $0.06 | per 1M tokens |
|
Llama 3.3
Meta Platforms
|
$0.52 | - | $0.52 | per 1M tokens |
|
MiniMax M2.5
MiniMax
|
$0.3 | $0.06 | $1.2 | per 1M tokens |
|
MiniMax M2.7
MiniMax
|
$0.15 | $0.06 | $0.6 | per 1M tokens |
|
MiniMax M3
MiniMax
|
$0.3 | $0.06 | $1.2 | per 1M tokens |
| $0.6 | $0.2 | $3.6 | per 1M tokens | |
|
Qwen 3.5 35B A3B
Alibaba
|
$0.6 | $0.35 | $3.6 | per 1M tokens |
|
Qwen 3.5 9B
Alibaba
|
$0.17 | - | $0.25 | per 1M tokens |
|
Qwen 3.6 Plus
Alibaba
|
$0.5 | - | $3 | per 1M tokens |
|
Qwen3 235B A22B
Alibaba
|
$0.1 | - | $0.3 | per 1M tokens |
|
Qwen3.7-Plus
Alibaba
|
$1.25 | $0.13 | $3.75 | per 1M tokens |
About Moonshot AI
Moonshot AI is a Beijing-based Chinese AI startup (founded 2023) focused on developing large language models and multimodal systems
View Company ProfileBenchmark scores
How Kimi K2.7 Code ranks on tracked AI benchmarks. Click any benchmark to see its full leaderboard.
Compare across all benchmarks →Tools using Kimi K2.7 Code
Last updated: July 9, 2026
