◆Franklin/Docs

Models

70+ models from 15+ providers, including a free tier.

Use a BlockRun API key for account billing: account setup guide. Register at user.blockrun.ai, create a key and add credits. Account mode needs no payment wallet; wallet billing supports Solana first, then Base.

Available Providers

Franklin gives you access to 70+ models from 15+ providers through a single interface. No API keys needed — everything is routed through BlockRun's unified gateway. Chat is billed at provider cost with no markup, plus $0.001 per call (media and live search: provider cost + 5%), settled instantly in USDC via x402.

  • Anthropic — Claude Fable 5.1 (1M context), Claude Opus 5.5 (1M context), Claude Sonnet 5.5 (1M context), Claude Fable 5, Claude Opus 5, Claude Opus 4.8, Claude Opus 4.7, Claude Sonnet 5, Claude Sonnet 4.6, Claude Haiku 4.5
  • OpenAI — GPT-6 Astra / Sol / Luna (1M context), GPT-5.6 Sol / Terra / Luna (plus Pro tiers), GPT-5.5, GPT-5.4, GPT-5.3 Codex, GPT-5.2, GPT-5.1, GPT-4.1, o1, o3
  • Google — Gemini 3.1 Pro, Gemini 3.6 Flash, Gemini 3.5 Flash, Gemini 3 Flash Preview, Gemini 2.5 Pro, Gemini 2.5 Flash, Flash Lite tiers
  • xAI — Grok 4.7, Grok 4.6 (500K context), Grok 4.5, Grok 4.3, Grok Build 0.1, Grok Imagine (image + video)
  • DeepSeek — DeepSeek V4 Pro (1M context), V4 Flash (chat + reasoner), V4 Flash Vision
  • Moonshot — Kimi K3 (1M context, vision + reasoning)
  • Z.AI — GLM-5.3, GLM-5.3 Flash (vision), GLM-5.2, GLM-5.1, GLM-5, GLM-5 Turbo
  • Qwen — Qwen3.7 Max (1M context), Qwen3.7 Plus, Qwen3.8 Flash (vision), Qwen3.7 Flash
  • MiniMax — MiniMax M3, MiniMax M2.7 (1M context, reasoning)
  • Xiaomi — MiMo V2.5 (vision), MiMo V2.5 Pro
  • NVIDIA, Cohere, Poolside — Nemotron 3, Llama 3.2 Vision, North Mini Code, Laguna XS (all free)

List All Models

See the full list of available models and their current status:

bash
franklin models

The output shows each model's name, provider, tier, and cost per million tokens.

Free Tier

The following models are always free — no wallet, no USDC, no sign-up:

  • Nemotron 3 Nano 30B (via NVIDIA — the fastest free model)
  • Nemotron 3.5 Lightning (via NVIDIA, 1M context)
  • Nemotron 3 Nano Omni (via NVIDIA, vision)
  • Nemotron 3 Ultra 550B (via NVIDIA, 1M context)
  • Llama 3.2 11B Vision (via NVIDIA)
  • Cohere North Mini Code (256K context, coding)
  • Poolside Laguna XS 2.1 (coding)
bash
# Use free models only
/model free

Free models are capable

The free NVIDIA models are surprisingly strong for most coding tasks. Try them before spending credits — they may be all you need.

Switching Models

Switch to a specific model mid-conversation with the /model command:

bash
# Switch to a specific model (shortcut or full model id)
/model sonnet
/model gpt-5.6-sol
/model anthropic/claude-opus-5.5
/model openai/gpt-6-sol
/model xai/grok-4.7

# Switch to a routing profile
/model auto
/model free

Model changes take effect immediately on the next message.

Prompt Caching

Anthropic models (Claude) support prompt caching, which reduces cost by approximately 75% for repeated context. This is enabled automatically — long conversations and system prompts are cached and reused across messages.

Caching is automatic

You don't need to configure anything. Franklin detects cacheable content and sends the appropriate cache headers to Anthropic's API automatically.