๐Ÿง 

Reasoning Council

Deep thinking models for complex analysis, logic puzzles, mathematical proofs, and research workloads. These models prioritize accuracy over speed.
๐Ÿ”ฌ Use case: Complex analysis, logic puzzles, research
7
Models
5
Providers
52
Max tok/s
Free
Best price

๐Ÿ“Š Council Models

ModelProviderSpeedโ˜…ContextPriceNotes
deepseek-v4-proOpenCode52 t/sโ˜…โ˜…โ˜…โ˜…โ˜…128KFlagship reasoning model
cogito-2.1:671bOllamaCloud45 t/sโ˜…โ˜…โ˜…โ˜…โ˜…128KFree671B params, massive model
dola-seed-2.0-proBytePlus50 t/sโ˜…โ˜…โ˜…โ˜…ยฝ64KBytePlus reasoning flagship
claude-opus-4-6Venice33 t/sโ˜…โ˜…โ˜…โ˜…โ˜…200KUncensored Claude Opus via Venice
claude-opus-4-6ORBIT29 t/sโ˜…โ˜…โ˜…โ˜…โ˜…200KFreeFree Claude Opus (2B tokens/mo)
qwen3-235b-thinkingChutes TEE28 t/sโ˜…โ˜…โ˜…โ˜…โ˜…128KThinking model, TEE-secured
deepseek-v3.2-teeChutes TEE28 t/sโ˜…โ˜…โ˜…โ˜…โ˜…128KDeep reasoning, TEE-secured

โš–๏ธ Pros & Cons

โœ… Pros

  • Claude Opus via ORBIT is free (2B tokens/mo)
  • TEE-secured options for confidential analysis
  • Thinking models (qwen3-235b, deepseek-v4-pro) show reasoning steps
  • 671B parameter model available (cogito-2.1)
  • Claude Opus has 200K context window

โš ๏ธ Cons

  • Slower generation speeds (28-52 t/s)
  • Premium models cost $9-20/mo
  • TEE models are slower than non-TEE equivalents
  • Large models may have longer time-to-first-token

๐Ÿš€ Try This Council

Send a complex reasoning prompt and compare deep analysis across models.

# Reasoning Council: Deep analysis with thinking model
curl -X POST https://api.opencode.ai/v1/chat/completions \
  -H "Authorization: Bearer $OPENCODE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-pro",
    "messages": [{"role": "user", "content": "Prove that the square root of 2 is irrational using proof by contradiction."}],
    "temperature": 0.1,
    "max_tokens": 2000
  }'