Models & pricing

Every hosted app on SkillSafe runs on one of these models. Usage is billed per token in credits — $1 = 10,000 credits — at the rates below, so you always know what a run costs before you ship or use an app.

Model Provider Input / 1M tokens Output / 1M tokens Max output Time limit
Claude Haiku 4.5 unavailable
claude-haiku-4-5
Anthropic $1.10 $5.50 4,096 tokens 60s
Claude Sonnet 5 unavailable
claude-sonnet-5
Anthropic $3.30 $16.50 8,192 tokens 120s
Claude Sonnet 4.6 unavailable
claude-sonnet-4-6
Anthropic $3.30 $16.50 8,192 tokens 120s
Claude Opus 4.8 unavailable
claude-opus-4-8
Anthropic $5.50 $27.50 8,192 tokens 180s
Claude Opus 5 unavailable
claude-opus-5
Anthropic $5.50 $27.50 8,192 tokens 180s
Claude Fable 5 unavailable
claude-fable-5
Anthropic $11.00 $55.00 16,384 tokens 180s
GPT-5.6 Sol
gpt-5.6-sol
OpenAI $5.50 $33.00 8,192 tokens 180s
GPT-5.6 Terra
gpt-5.6-terra
OpenAI $2.75 $16.50 8,192 tokens 120s
GPT-5.6 Luna
gpt-5.6-luna
OpenAI $1.10 $6.60 4,096 tokens 60s
GPT-5.1
gpt-5.1
OpenAI $1.38 $11.00 8,192 tokens 120s
GPT-5 mini
gpt-5-mini
OpenAI $0.28 $2.20 4,096 tokens 60s
Gemma 4 26B (Workers AI) default
@cf/google/gemma-4-26b-a4b-it
Workers AI $0.11 $0.33 4,096 tokens 60s
GPT Image 2 image
gpt-image-2
OpenAI ~$0.066 / image 1 image 180s
FLUX.2 Klein 4B (Workers AI) image
@cf/black-forest-labs/flux-2-klein-4b
Workers AI ~$0.0016 / image 1 image 60s

Rates are provider list price plus a 10% platform margin, which covers Stripe fees, refunds, and infrastructure. A creator's markup is a separate cut of the same base — they keep 100% of it.

How pricing works

  • Pay for what a run actually uses. Each run is charged on the tokens the model actually consumed, rounded up to whole credits (minimum 1 credit ≈ $0.0001), plus a fixed per-job overhead of $0.001. On the cheapest models that overhead is most of what a run costs — it covers the queue, storage and ledger work every run does regardless of model.
  • Two equal cuts, both on the same base. A run's base is the provider's list cost plus the $0.001 overhead. The platform takes 10% of it; the creator takes their markup — up to 100% — of the very same base, shown on each app's detail page. At the default 10% the two are equal, and nothing is deducted from the creator's side: they keep 100% of their markup.
  • BYOK apps bill a flat 1 credit per run. When a publisher brings their own provider key, their key pays for inference and the platform charges only a 1-credit overhead — markup is forced to zero.
  • Spend is capped per job. Every model carries hard caps on input tokens, output tokens, and wall-clock time, so a single run can never overrun its hold.
  • Image models bill per image. An image-generation run produces one 1024×1024 image. The hold reserves the worst-case cost of that image and the run settles down to the provider's actual (usage-reported where available) cost — same base, same two cuts as text runs.
  • No model configured? Apps without a model run on the Workers AI default — cheap, keyless, and never silently billed at premium rates.