NOUMENOpen app
NOUMEN

longcat · Open weights

LongCat 2.0

Added Jun 30, 2026Open weights

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

Use in the app

Meituan's LongCat 2.0 is an open-weight agentic model for coding, tool use, and long-context workflow automation, with reasoning disabled for faster direct answers.

  • Tools
Input
Text
Output
Text
Discovery tier
Open weights

Pricing

USD per 1M tokens
Input
$0.75
Output
$3.00
Cache read
$0.015

Reference prices as published by the model provider. API example

Specifications

Context window
1M
Max output
131.1K
Average output (7 days)
368 tokens
Longer than 38% of 468 models

Side by side

LongCat 2.0 compared

Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.

Compare models

Choose up to 3 models to compare.1 of 3 chosen

Loading models…

LongCat 2.0longcatThis modelGLM 4.6 Turbo (Thinking)Z.aiRemove
ProviderlongcatZ.ai
Discovery tierOpen weightsOpen weights
AddedJun 30, 2026Oct 2, 2025
Limits
Context window1MBest204.8K
Max output131.07K131.07K
Average output368 tokens1.9K tokens
Pricing, USD per 1M tokens
Input$0.75Best$1.00
Output$3.00$3.00
Cache read$0.015—
Performance
Request success (24h)100.0%—
Benchmark indexes
Intelligence Index19.1Best18.5
Coding Index45.345.8Best
Math Index—86.0
Capabilities
ReasoningNoYes
ToolsYesNo
Reasoning effort——

Public evaluations

Benchmarks

Independent evaluation results, with how each score ranks among the models in the Noumen catalog.

  • LongCat 2.0
  • GLM 4.6 Turbo (Thinking)

Intelligence Index

  • LongCat 2.019.1
  • GLM 4.6 Turbo (Thinking)18.5

LongCat 2.0: Better than 48% of 326 scored models

3.8Catalog models by score55.1

Indexes

  • Coding Index

    Composite of the coding evaluations

    • LongCat 2.0: 45.3
    • GLM 4.6 Turbo (Thinking): 45.8Top

  • Math Index

    Composite of the math evaluations

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 86.0

Agentic work

  • AutomationBench-AA

    Workflow automation, penalized for guardrail violations

    • LongCat 2.0: 4.5%
    • GLM 4.6 Turbo (Thinking): —

  • AutomationBench-AA tasks completed

    Workflows finished end to end without a guardrail violation

    • LongCat 2.0: 0.5%
    • GLM 4.6 Turbo (Thinking): —

  • AA-Briefcasev1.1

    Agentic knowledge work, rated head to head

    • LongCat 2.0: 739 Elo
    • GLM 4.6 Turbo (Thinking): —

  • GDPval-AAv2.1

    Economically valuable occupational tasks, rated head to head

    • LongCat 2.0: 869 EloTop
    • GLM 4.6 Turbo (Thinking): 756 Elo

Document reasoning

  • GDP.pdf

    Professional reasoning over PDF documents, all-pass rate

    • LongCat 2.0: 6.4%
    • GLM 4.6 Turbo (Thinking): —

  • AA-LCRv1.1

    Reasoning across long documents

    • LongCat 2.0: 65.0%Top
    • GLM 4.6 Turbo (Thinking): 54.0%

Reasoning and knowledge

  • Humanity’s Last Exam

    Expert-level questions across academic fields

    • LongCat 2.0: 33.7%Top
    • GLM 4.6 Turbo (Thinking): 14.5%

  • GPQA Diamond

    Graduate-level science questions

    • LongCat 2.0: 78.0%Top
    • GLM 4.6 Turbo (Thinking): 78.0%Top

  • MMLU-Pro

    Multi-subject knowledge and reasoning

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 82.9%

  • IFBench

    Following precise output instructions

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 43.4%

Coding

  • Terminal-Benchv4.0

    Practical software engineering in a terminal

    • LongCat 2.0: 0.0%
    • GLM 4.6 Turbo (Thinking): —

  • SciCode

    Scientific computing in Python

    • LongCat 2.0: 36.3%
    • GLM 4.6 Turbo (Thinking): —

  • LiveCodeBench

    Recent competitive programming problems

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 69.5%

Math

  • AIME 2025

    Competition mathematics, 2025 problems

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 86.0%

Legacy evaluations

  • Terminal-Bench Hard

    Earlier terminal task suite

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 25.0%

  • τ²-Bench Telecom

    Tool use in a customer-support simulation

    • LongCat 2.0: —
    • GLM 4.6 Turbo (Thinking): 70.5%

  • AA-LCR (unversioned)

    Long-context reasoning, earlier grading

    • LongCat 2.0: 65.0%Top
    • GLM 4.6 Turbo (Thinking): 54.0%

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.

Request success (24h)
100.0%
Reliable
Hourly request success · last 72 hoursLowest hour 97.6%
  • 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.

Read the quickstart
curl https://www.noumen.xyz/api/v1/chat/completions \
  -H "Authorization: Bearer $NOUMEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "longcat-2.0",
    "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="longcat-2.0",
    messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)

longcat

Other longcat models, starting with the same family.