Moonshot · Open weights
Kimi K2 0905
Kimi K2 0905. Kimi-k2 is a Mixture-of-Experts (MoE) foundation model with exceptional coding and agent capabilities, featuring 1 trillion total parameters and 32 billion activated parameters. In benchmark evaluations covering general knowledge reasoning, programming, mathematics, and agent-related tasks, the K2 model outperforms other leading open-source models. Quantized at FP8.
- Tools
- Structured output
- Input
- Text
- Output
- Text
- Discovery tier
- Open weights
Pricing
USD per 1M tokens- Input
- $0.40
- Output
- $1.80
Reference prices as published by the model provider. API example
Specifications
- Context window
- 262.1K
- Max output
- 100.4K
- Average output (7 days)
- 677 tokens
- Longer than 60% of 465 models
Side by side
Kimi K2 0905 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Kimi K2 0905MoonshotThis model | Kimi K3 TEEMoonshotRemove | |
|---|---|---|
| Provider | Moonshot | Moonshot |
| Discovery tier | Open weights | Open weights |
| Added | Sep 25, 2025 | Jul 31, 2026 |
| Limits | ||
| Context window | 262.14K | 1.05MBest |
| Max output | 100.35KBest | 65.54K |
| Average output | 677 tokens | 1.1K tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.40Best | $3.00 |
| Output | $1.80Best | $15.00 |
| Cache read | — | $1.50 |
| Performance | ||
| Throughput | 50.8 tok/sBest | 31.7 tok/s |
| Time to first token | 3.3 sBest | 5.0 s |
| Request success (24h) | 100.0%Best | 98.0% |
| Benchmark indexes | ||
| Intelligence Index | 15.3 | 43.6Best |
| Coding Index | — | 76.2 |
| Math Index | 57.3 | — |
| Capabilities | ||
| Reasoning | No | Yes |
| Tools | Yes | Yes |
| Vision | No | Yes |
| Structured output | Yes | Yes |
| Reasoning effort | — | low, high, max |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Kimi K2 0905
- Kimi K3 TEE
Intelligence Index
- Kimi K2 090515.3
- Kimi K3 TEE43.6
Kimi K2 0905: Better than 39% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Kimi K2 0905: —
- Kimi K3 TEE: 76.2
Math Index
Composite of the math evaluations
- Kimi K2 0905: 57.3
- Kimi K3 TEE: —
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Kimi K2 0905: —
- Kimi K3 TEE: 58.3%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Kimi K2 0905: —
- Kimi K3 TEE: 24.8%
Harvey LAB-AAv1.1
Legal tasks passed in full, with no material hallucination
- Kimi K2 0905: —
- Kimi K3 TEE: 5.3%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Kimi K2 0905: —
- Kimi K3 TEE: 1,499 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Kimi K2 0905: —
- Kimi K3 TEE: 1,534 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Kimi K2 0905: —
- Kimi K3 TEE: 22.0%
AA-LCRv1.1
Reasoning across long documents
- Kimi K2 0905: 53.0%
- Kimi K3 TEE: 88.7%Top
MLCR-AA
Long-context reasoning over medical records
- Kimi K2 0905: —
- Kimi K3 TEE: 21.7%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Kimi K2 0905: 6.4%
- Kimi K3 TEE: 46.9%Top
GPQA Diamond
Graduate-level science questions
- Kimi K2 0905: 76.7%
- Kimi K3 TEE: 93.5%Top
MMLU-Pro
Multi-subject knowledge and reasoning
- Kimi K2 0905: 81.9%
- Kimi K3 TEE: —
IFBench
Following precise output instructions
- Kimi K2 0905: 41.7%
- Kimi K3 TEE: —
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Kimi K2 0905: —
- Kimi K3 TEE: 12.6%
SciCode
Scientific computing in Python
- Kimi K2 0905: —
- Kimi K3 TEE: 59.5%
LiveCodeBench
Recent competitive programming problems
- Kimi K2 0905: 61.0%
- Kimi K3 TEE: —
Math
AIME 2025
Competition mathematics, 2025 problems
- Kimi K2 0905: 57.3%
- Kimi K3 TEE: —
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- Kimi K2 0905: 23.5%
- Kimi K3 TEE: —
τ²-Bench Telecom
Tool use in a customer-support simulation
- Kimi K2 0905: 73.4%
- Kimi K3 TEE: —
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Kimi K2 0905: 53.0%
- Kimi K3 TEE: 88.7%Top
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Throughput
- 50.8 tok/s
- Faster than 38% of 250 models
- Time to first token
- 3.3 s
- Quicker than 52% of 250 models
- Prompt cache hits
- 67.5%
- Request success (24h)
- 100.0%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "moonshotai/Kimi-K2-Instruct-0905",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="moonshotai/Kimi-K2-Instruct-0905",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)Moonshot
Related models
Other Moonshot models, starting with the same family.