Simple, Transparent Pricing
Start free. Scale when you need to.
Free + BYOK
$0/forever
Bring your own API keys — pay only providers
Get Started Free
- 5 debates/week, 20/month
- 2 models per debate
- 3 basic deliberation modes
- 3 starter templates
- Streaming support
- Basic audit trail
- Basic analytics (30-day history)
- Community support
- Markdown export
- All models via BYOK
- Wallet top-up available
Coming Soon
Pro
$29/mo
or $290/yr (2 months free)
For developers who use multi-model debate daily
Coming Soon
- Unlimited deliberations
- 5+ models per debate
- All 8 deliberation modes
- All 8 templates + custom creation
- 500K weighted tokens/week included
- Full CLI with codebase context
- MCP integration
- SDK access (TypeScript & Python)
- Full audit trail
- Advanced analytics + diversity scores
- Unlimited history + full-text search
- Weekly email digest
- Priority email support
- JSON + Markdown export
- 99.9% SLA
Coming Soon
Max
$99/mo
or $990/yr (2 months free)
Verified deliberation with research-backed trust features
Coming Soon
- Everything in Pro
- 2M weighted tokens/week included
- Unlimited models per debate
- Parallel & batch CLI mode
- REST API access for automation
- Conformal safety gate (81.9% wrong consensus caught)
- Progressive deepening (80-92% token savings)
- Debate collapse detection
- Disagreement map visualization
- Argument graph (interactive)
- Persistent project memory
- Decision audit trail (searchable)
- Daily + weekly digest options
- Direct founder support
- PDF + JSON + Markdown export
Detailed Feature Comparison
| Feature | Free + BYOK | Pro | Max |
|---|---|---|---|
| Weekly compute | 50K weighted tokens | 500K weighted tokens | 2M weighted tokens |
| Debates | 5/week, 20/month | Unlimited | Unlimited |
| Models per debate | 2 | 5+ | Unlimited |
| Deliberation modes | 3 basic | All 8 | All 8 + early access |
| Templates | 3 starter | All 8 + custom | All + share publicly |
| CLI | Basic | Full + codebase context | Full + parallel/batch |
| MCP | Full + streaming | ||
| SDK access | |||
| REST API | |||
| BYOK (own API keys) | |||
| Wallet top-up | |||
| On-demand charging | |||
| Streaming | |||
| Audit trail | Basic | Full | Full + searchable |
| Analytics | Basic | Advanced | Advanced + replay |
| History | 30 days | Unlimited + search | Unlimited + API |
| Conformal safety gate | |||
| Progressive deepening | |||
| Debate collapse detection | |||
| Disagreement map | |||
| Argument graph | |||
| Persistent memory | |||
| Support | Community | Priority email | Founder access |
| SLA | None | 99.9% | 99.9% |
Model Cost Reference
These are provider costs per 1M tokens, not Consilium charges. Consilium adds zero markup.
| Provider | Model | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| Anthropic | Claude Opus 4.7 | $5.00 | $25.00 |
| Claude Opus 4.6 | $5.00 | $25.00 | |
| Claude Sonnet 4.6 | $3.00 | $15.00 | |
| Claude Haiku 4.5 | $1.00 | $5.00 | |
| OpenAI | GPT-5.5 Pro | $8.00 | $32.00 |
| GPT-5.5 | $3.00 | $12.00 | |
| GPT-5.4 | $2.00 | $8.00 | |
| GPT-5.4 Mini | $0.20 | $0.80 | |
| GPT-5.4 Nano | $0.08 | $0.30 | |
| Gemini 3.1 Pro | $1.25 | $5.00 | |
| Gemini 3 Flash | $0.15 | $0.60 | |
| Gemini 3.1 Flash-Lite | $0.05 | $0.20 | |
| Groq | Llama 3.1 8B Instant | Free* | Free* |
| Llama 3.3 70B Versatile | Free* | Free* | |
| GPT-OSS 120B (via Groq) | $0.15 | $0.60 | |
| GPT-OSS 20B (via Groq) | $0.05 | $0.15 | |
| Groq Compound | $0.80 | $1.60 | |
| Groq Compound Mini | $0.30 | $0.60 | |
| xAI | Grok 4.20 | $3.00 | $15.00 |
| Grok 4.1 Fast (reasoning) | $1.00 | $4.00 | |
| Grok 4.1 Fast (non-reasoning) | $0.50 | $2.00 | |
| Grok Code Fast | $0.30 | $1.20 | |
| Moonshot | Kimi K2.6 | $1.20 | $2.50 |
| OpenRouter (free tier) | Gemma 4 26B (free) | Free | Free |
| Gemma 4 31B (free) | Free | Free | |
| Qwen3 Coder (free) | Free | Free | |
| Nemotron 3 Super 120B (free) | Free | Free | |
| Ling 2.6 1T (free) | Free | Free |
Cost Examples
Quick question with GPT-5.4 Mini~$0.001
1 round, 1 modelCouncil mode with 3 premium models~$0.05 - $0.15
3 rounds, 3 modelsDeep mode with 5 models~$0.20 - $0.50
5 rounds, 5 models + sub-agentsAny mode with Groq models$0.00
Llama models are freeCost Estimation Formula
estimated_cost = num_api_calls * estimated_tokens * cost_per_tokenQuick
1 call per modelCouncil
num_models * 3 roundsDeep
num_models * 5 roundsRed Team
num_models * 3 phasesJury
num_models * 3 rounds + dissentMarket
num_models * 2 rounds + aggregationNeed a custom arrangement?
Volume pricing, specific compliance requirements, or a private deployment — talk to us. We work with teams individually until Consilium has the traction for a formal Enterprise tier.
Talk to us