Why do I get a 401 Unauthorized error?

Check if your AgentLLM key is valid and correctly passed in the Authorization header. You can regenerate the key under Test → LLM → Keys.

Why is the wrong version of my agent responding?

Check the 'model' string in your request. Make sure you're using the correct version tag like @dev, @prod, or @1.0.

What do I do if my request returns no output?

Confirm that your agent is correctly configured and LLM is toggled on. Then inspect logs under Deployments → Logs.

How do I find the correct baseURL and model string?

Go to Test → LLM → Code. Always copy the baseURL and model directly from there to avoid typos or environment mismatches.

Billing Management

Use this page to check per-unit pricing for models, services, and components in SmythOS. This helps you forecast costs, avoid surprises, and use credits wisely. You can also view subscription plans for limits and credit multipliers.

Using SmythOS-managed models

You don’t need API keys for default models. SmythOS manages setup, billing, and usage limits.

Billing Principles

Real-time usage insights

Use the Analytics Dashboard to view task activity, token consumption, and exports, and set limits on the My Plan page.

Each plan includes monthly credits
Overages are billed pay-as-you-go
Charges apply only after credits are exhausted
External API model charges are non-refundable
Failed runs or tests in development are never billed

Model Pricing

Need plan-level limits?

This section covers unit pricing. For plan quotas and features, see Compare Plans and Features.

OpenAI Models

Model	Input Price (per 1M tokens)	Cached Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Search Price (per 1k calls)	Image Support
GPT-5	$1.30	$0.13	$10.00	400k tokens	$10.00	Text only (token-based)
GPT-5 mini	$0.30	$0.03	$2.00	400k tokens	$10.00	Text only (token-based)
GPT-5 nano	$0.10	$0.01	$0.40	400k tokens	N/A	Text only (token-based)
GPT-5 Chat	$1.30	$0.13	$10.00	400k tokens	N/A	Text only (token-based)

Search Pricing Notes

Models with N/A in the Search Price column do not currently support the Search API in SmythOS.

View Legacy OpenAI Models

Warning: These models are legacy in SmythOS. They are not recommended for new projects and may lose support over time.

Model	Input Price (per 1M tokens)	Cached Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Search Price (per 1k calls)	Image Support	Status
GPT-4o	$2.50	$1.25	$25.00	128k tokens	$25.00	Input only (token-based)	Legacy
GPT-4o mini	$0.15	$0.075	$0.60	128k tokens	$25.00	Input only	Legacy
GPT-4.1	$2.00	$0.50	$8.00	1M tokens	$25.00	Input only	Discontinued
GPT-4.1 mini	$0.40	$0.10	$1.60	1M tokens	$25.00	Input only	Discontinued
GPT-4.1 nano	$0.10	$0.025	$0.40	1M tokens	N/A	Input only	Discontinued
o3	$1.10	$0.55	$4.40	200k tokens	N/A	Input only	Legacy
o4-mini	$0.10	$0.05	$0.30	200k tokens	N/A	Input only	Discontinued

Image token calculation

For models that accept image input, pricing is based on text tokens.
Approximate tokens = (width px × height px) / 750.

Gemini (Google AI)

Model	Search Price (per 1k calls)	Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Media Pricing
Gemini 2.5 Flash	$0.035	$0.10	$0.40	1M tokens	$0.70/image, $0.10/media
Gemini 2.5 PRO ≤200k	$0.035	$1.25	$10.00	1M tokens	$0.70/image, $0.10/media
Gemini 2.5 PRO >200k	$0.035	$2.50	$15.00	1M tokens	$0.70/image, $0.10/media

Claude (Anthropic)

Model	Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Notes
Claude Sonnet 4	$3.00	$15.00	200k tokens	High-performance version
Claude Opus 4	$15.00	$75.00	200k tokens	High-performance reasoning
Claude Opus 4.1	$15.00	$75.00	200k tokens	Most capable model yet, excels at complex reasoning & advanced coding

View Legacy Claude Models

Note: These models are legacy in SmythOS. They are still available, but not recommended for new projects.

Model	Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Status
Claude 3 Opus	$15.00	$75.00	200k tokens	Legacy
Claude 3.5 Haiku	$0.80	$4.00	200k tokens	Legacy
Claude 3.5 Sonnet Latest	$3.00	$15.00	200k tokens	Legacy
Claude 3.5 Sonnet Stable	$3.00	$15.00	200k tokens	Legacy
Claude 3.7 Sonnet	$3.00	$15.00	200k tokens	Legacy
Claude 3.7 Sonnet Thinking	$3.00	$15.00	200k tokens	Legacy

Perplexity Models

Model	Search Price (per 1k calls)	Input Price (per 1M tokens)	Output Price (per 1M tokens)	Max Context Size	Notes
Sonar	$0.005	$1.00	$1.00	128k tokens	Basic model
Sonar PRO	$0.005	$3.00	$15.00	200k tokens	Premium tier
Sonar Reasoning Pro	$0.005	$2.00	$8.00	128k tokens	Reasoning-optimized
Sonar Deep Research	$0.005	$2.00	$8.00	128k tokens	+$3.00 for deep reasoning

Core Services and Components

Execution Costs

Service	Rate
Weaver Engine	$3.00 input / $12.00 output per 1M tokens
Task Execution	$0.001 per completed task
NodeJS Runtime	$0.0001 per second
Vanilla JS	Free

What counts as a task?

1 component = 1 task
Bulk run with 100 rows × 2 components = 200 tasks
Failed runs = 0 tasks
Development/test runs = Free

Production-only pricing

Task costs only apply in live deployments. Testing in the builder is always free.

Image Tools

Tool	Provider	Rate
GPT Image	OpenAI	$5 input, $10 image input, $40 output per 1M tokens
HD 8 Steps	Flux Schnell	$0.001 per request
HD 50 Steps	Flux Dev	$0.008 per request

Web Utilities

Tool	Provider	Cost
Web Search	Tavily	$0.008 per search
Web Scraping	ScrapFly	$0.00015 per credit

System Components

Component	Cost
NodeJS Runtime	$0.0001 per second
Computer Use	$0.0025 per second
MCP	Model-specific pricing

Controlling Costs

Was this page helpful?

Billing Principles​

Model Pricing​

OpenAI Models​

Gemini (Google AI)​

Claude (Anthropic)​

Perplexity Models​

Core Services and Components​

Execution Costs​

Image Tools​

Web Utilities​

System Components​

Controlling Costs​