Moonshot · Open weights
Kimi K2.6
Kimi K2.6 is an open-source, native multimodal agentic model built for long-horizon coding, coding-driven design, and large-scale task orchestration. It can turn simple prompts and visual inputs into production-ready interfaces and full-stack workflows, and is designed to coordinate complex multi-agent plans with thousands of steps across code, documents, and spreadsheets.
- Tools
- Vision
- Input
- Text, Images
- Output
- Text
- Discovery tier
- Open weights
Pricing
USD per 1M tokens- Input
- $0.45
- Output
- $1.89
- Cache read
- $0.0757
Reference prices as published by the model provider. API example
Specifications
- Context window
- 262.1K
- Max output
- 65.5K
- Average output (7 days)
- 594 tokens
- Longer than 55% of 465 models
Side by side
Kimi K2.6 compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Kimi K2.6MoonshotThis model | GLM 5 TurboZ.aiRemove | |
|---|---|---|
| Provider | Moonshot | Z.ai |
| Discovery tier | Open weights | Open weights |
| Added | Apr 16, 2026 | Mar 15, 2026 |
| Limits | ||
| Context window | 262.14KBest | 202.8K |
| Max output | 65.54K | 131.07KBest |
| Average output | 594 tokens | 557 tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.45Best | $1.20 |
| Output | $1.89Best | $4.00 |
| Cache read | $0.0757Best | $0.24 |
| Performance | ||
| Throughput | 46.6 tok/s | — |
| Time to first token | 2.7 s | — |
| Request success (24h) | 99.7% | 100.0%Best |
| Benchmark indexes | ||
| Intelligence Index | 27.0Best | 26.6 |
| Coding Index | 61.8 | — |
| Capabilities | ||
| Tools | Yes | Yes |
| Vision | Yes | No |
| Structured output | No | Yes |
| Reasoning effort | — | — |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Kimi K2.6
- GLM 5 Turbo
Intelligence Index
- Kimi K2.627.0
- GLM 5 Turbo26.6
Kimi K2.6: Better than 68% of 326 scored models
Indexes
Coding Index
Composite of the coding evaluations
- Kimi K2.6: 61.8
- GLM 5 Turbo: —
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Kimi K2.6: 13.0%
- GLM 5 Turbo: —
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Kimi K2.6: 1.8%
- GLM 5 Turbo: —
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Kimi K2.6: 819 Elo
- GLM 5 Turbo: —
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Kimi K2.6: 1,038 Elo
- GLM 5 Turbo: —
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Kimi K2.6: 13.0%
- GLM 5 Turbo: —
AA-LCRv1.1
Reasoning across long documents
- Kimi K2.6: 81.0%Top
- GLM 5 Turbo: 71.7%
MLCR-AA
Long-context reasoning over medical records
- Kimi K2.6: 11.1%
- GLM 5 Turbo: —
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Kimi K2.6: 37.5%Top
- GLM 5 Turbo: 27.8%
GPQA Diamond
Graduate-level science questions
- Kimi K2.6: 91.1%Top
- GLM 5 Turbo: 84.7%
IFBench
Following precise output instructions
- Kimi K2.6: 76.0%Top
- GLM 5 Turbo: 73.2%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Kimi K2.6: 0.5%
- GLM 5 Turbo: —
SciCode
Scientific computing in Python
- Kimi K2.6: 51.5%
- GLM 5 Turbo: —
Legacy evaluations
Terminal-Bench Hard
Earlier terminal task suite
- Kimi K2.6: 43.9%Top
- GLM 5 Turbo: 33.3%
τ²-Bench Telecom
Tool use in a customer-support simulation
- Kimi K2.6: 95.9%
- GLM 5 Turbo: 98.5%Top
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Kimi K2.6: 81.0%Top
- GLM 5 Turbo: 71.7%
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
- Kimi K2.6
- GLM 5 Turbo
| Leaderboard | Kimi K2.6 | GLM 5 Turbo |
|---|---|---|
| FrontierMath v2 (Tiers 1–3) | #8 · 57.2% | Not in top 20 |
| SWE-bench Verified | #5 · 76.7% | Not in top 20 |
| GPQA Diamond | #9 · 90.8% | Not in top 20 |
Ranks among Noumen catalog models on Epoch AI leaderboards. Only top-20 placements are listed.Epoch AI
- Kimi K2.6
- GLM 5 Turbo
| Leaderboard | Kimi K2.6 | GLM 5 Turbo |
|---|---|---|
| EQ-Bench 4 | #20 · 1194 | Not in top 20 |
| Longform Writing | #19 · 78.5 | Not in top 20 |
| EQ-Bench 3 | #10 · 1548 | Not in top 20 |
Ranks among Noumen catalog models on EQ-Bench leaderboards. Only top-20 placements are listed.EQ-Bench
- Kimi K2.6
- GLM 5 Turbo
| Leaderboard | Kimi K2.6 | GLM 5 Turbo |
|---|---|---|
| Agentic Game Dev | #17 · 1105 | Not in top 20 |
| Native Apps | #16 · 1163 | Not in top 20 |
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
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
- 46.6 tok/s
- Faster than 34% of 250 models
- Time to first token
- 2.7 s
- Quicker than 61% of 250 models
- Prompt cache hits
- 54.9%
- About $0.24 per 1M input tokens with caching
- Request success (24h)
- 99.7%
- 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-k2.6",
"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-k2.6",
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.