Cogito v2 preview - 671B MoE

World-class MoE reasoning approaching superintelligence

About model

Cogito 671B MoE represents one of the strongest open models globally, matching performance of latest Deepseek models while approaching closed frontier systems like o3 and Claude 4 Opus. This advanced system demonstrates significant progress toward scalable superintelligence through policy improvement.

Quickstart guides

Agents

Reasoning quickstart

Apps

Structured outputs Docs

Infrastructure

Accelerate LLM Inference (Guide)

Performance benchmarks

Model	GPQA Diamond	HLE	LiveCodeBench	MATH500	SWE-bench verified
Cogito v2 preview - 671B MoE	69.7%
Related open-source models
Competitor closed-source models
Claude Opus 4.6	90.5%	34.2%			78.7%
OpenAI o3	83.3%	24.9%		99.2%	62.3%
OpenAI o1	76.8%			96.4%	48.9%
GPT-4o	49.2%	2.7%	32.3%	89.3%	31.0%

API usage

cURL
Python
Typescript

Endpoint:

deepcogito/cogito-v2-preview-deepseek-671b

curl -X POST "https://api.together.xyz/v1/chat/completions" \
  -H "Authorization: Bearer $TOGETHER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepcogito/cogito-v2-preview-deepseek-671b",
    "messages": [
      {
        "role": "user",
        "content": "What are some fun things to do in New York?"
      }
    ]
}'

from together import Together

client = Together()

response = client.chat.completions.create(
  model="deepcogito/cogito-v2-preview-deepseek-671b",
  messages=[
    {
      "role": "user",
      "content": "What are some fun things to do in New York?"
    }
  ]
)
print(response.choices[0].message.content)

import Together from 'together-ai';
const together = new Together();

const completion = await together.chat.completions.create({
  model: 'deepcogito/cogito-v2-preview-deepseek-671b',
  messages: [
    {
      role: 'user',
      content: 'What are some fun things to do in New York?'
     }
  ],
});

console.log(completion.choices[0].message.content);

Model card

This is a hybrid reasoning model. To enable thinking mode, pass the following parameter with your request to the model:


    
"chat_template_kwargs": {"enable_thinking": true}

Here's an example cURL request with thinking enabled:

    
curl -X POST "https://api.together.xyz/v1/chat/completions" \
  -H "Authorization: Bearer $TOGETHER_API_KEY" \
  -H "Content-Type: application/json" \
    -d '{
    "model": "deepcogito/cogito-v2-preview-deepseek-671b",
    "temperature": 0.6,
    "chat_template_kwargs": {"enable_thinking": true},
    "messages": [
      {
        "role": "user",
        "content": "What are some fun things to do in New York?"
      }
    ]
}'

Here's an example Python request with thinking enabled:

    
cfrom together import Together

client = Together()

chat_template_kwargs = {"enable_thinking": True}

response = client.chat.completions.create(
  model="deepcogito/cogito-v2-preview-deepseek-671b",
  extra_body={"chat_template_kwargs": chat_template_kwargs},
  messages=[
    {
      "role": "user",
      "content": "What are some fun things to do in New York?"
    }
  ]
)
print(response.choices[0].message.content)

‍
Architecture Overview:
• Massive 671B mixture-of-experts architecture with intelligent routing
• World-class reasoning capabilities among strongest open models
• Advanced policy improvement for both reasoning and non-reasoning modes

Training Methodology:
• Dual-mode training improving both standard and reasoning performance
• Signal-based training for thinking process optimization
• Advanced distillation techniques preventing reasoning meandering

Performance Characteristics:
• Matches Deepseek v3 0324 in non-reasoning mode
• Outperforms Deepseek R1 with 60% shorter reasoning chains
• Approaches performance of o3 and Claude 4 Opus frontier models

Applications & use cases
Frontier Research:
• Advanced scientific research and discovery
• Complex theoretical analysis and mathematical proofs
• Multi-disciplinary research requiring world-class reasoning

Strategic Applications:
• High-stakes decision making and strategic planning
• Advanced competitive analysis and market research
• Complex system design and optimization

Superintelligence Development:
• Foundation for next-generation AI research
• Scalable self-improvement research and development
• Open source contribution to AGI and superintelligence efforts
‍

Related models

Model specifications

Model data

Model provider
Deep Cogito
Type
Chat
Reasoning
Main use cases
Chat
Reasoning
Features
JSON Mode
Deployment
Serverless
On-Demand Dedicated
Monthly Reserved
Endpoint
deepcogito/cogito-v2-preview-deepseek-671b
Parameters
671B MoE
Input price
$1.25 / 1M tokens
Output price
$1.25 / 1M tokens
Input modalities
Text
Output modalities
Text

Category
Chat

Run in Playground

Quickstart docs

Deploy model

Cogito v2 preview - 671B MoE

About model

API usage

Model card

Applications & use cases