Compare

Kimi K3 vs MiniMax M3

Pricing per million tokens, context window, capabilities — pulled from each provider's public docs. All 2 are available via the same AIgateway OpenAI-compatible endpoint; flip the model string to switch.

Search2/4
Kimi K3
moonshot/kimi-k3
MiniMax M3
minimax/m3
Provider
Moonshot
MiniMax
Family
MiniMax M
Modality
text
text
Context window
1,048,576 tok
1,000,000 tok
Max output
32,768 tok
4,096 tok
Released
2026-07-16
2026-06-01
License
Open-weight
Open-weight
Input price
$3.00 /1M
$0.300 /1M
Output price
$15.00 /1M
$1.20 /1M
Tools
yes
Streaming
yes
yes
Vision
yes
JSON mode
yes
Reasoning
yes
Prompt caching
Batch API
Try it
Open in playground →
Open in playground →
Kimi K3
moonshot/kimi-k3
Full spec →

Kimi K3 is Moonshot's flagship 2.8 trillion-parameter model, built on Kimi Delta Attention (a hybrid linear attention mechanism) with Attention Residuals. It offers native visual understanding, always-on reasoning, and a 1M-token context window for long-horizon coding, knowledge work, and deep reasoning tasks.

Strengths
  • General-purpose chat
  • Long context
  • Tool use
Use cases
ChatbotsContent generationAgentic workflows
MiniMax M3
minimax/m3
Full spec →

MiniMax's M3 language model with frontier coding and agentic capabilities, a 1M token context window, and multilingual support.

Strengths
  • General-purpose chat
  • Long context
  • Tool use
Use cases
ChatbotsContent generationAgentic workflows

Compare with another

Claude Fable 5.1 vs Kimi K3
anthropic/claude-fable-5.1 · moonshot/kimi-k3
Claude Fable 5.1 vs MiniMax M3
anthropic/claude-fable-5.1 · minimax/m3
Gemini 3.8 Flash vs Kimi K3
google/gemini-3.8-flash · moonshot/kimi-k3
Gemini 3.8 Flash vs MiniMax M3
google/gemini-3.8-flash · minimax/m3
Claude Opus 5 vs Kimi K3
anthropic/claude-opus-5 · moonshot/kimi-k3
Claude Opus 5 vs MiniMax M3
anthropic/claude-opus-5 · minimax/m3
Grok 4.6 vs Kimi K3
xai/grok-4.6 · moonshot/kimi-k3
Grok 4.6 vs MiniMax M3
xai/grok-4.6 · minimax/m3
Glm-5.3 vs Kimi K3
zai-org/glm-5.3 · moonshot/kimi-k3
SWITCH BETWEEN THEM

One key, all 2, one line different.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.aigateway.sh/v1",
    api_key="sk-aig-...",
)

# Kimi K3
client.chat.completions.create(
    model="moonshot/kimi-k3",
    messages=[{"role":"user","content":"hello"}],
)

# MiniMax M3
client.chat.completions.create(
    model="minimax/m3",
    messages=[{"role":"user","content":"hello"}],
)
Get an AIgateway keyRun an eval on these →