Digital Mechanics Digital Mechanics Laboratory
ruen

AI model pricing

Prices are in coins per 1 million tokens.

Claude Fable 5Maximum quality for complex architecture, deep analysis and tasks where the cost of a mistake is especially high.Flagship11,00070,0001,30018,000
Claude Opus 5For large refactorings, ambiguous requirements and thorough code review.Flagship5,20035,0002509,000
GPT 5.6 Sol ProFor the most complex reasoning, architectural decisions and mission-critical code.Flagship5,00040,0006009,000
Claude Sonnet 5The main strong choice for development, complex logic and large edits without the flagship price tag.Advanced2,50015,0003504,000
GPT 5.6 Terra ProA strong trade-off between development and analysis without the top-end cost.Advanced2,50024,0003504,000
Gemini 3.1 ProFor multi-step analysis, planning and working with a large context.Advanced2,20018,0002203,500
Grok 4.5For research tasks, weighing alternatives and unconventional solutions.Advanced2,10010,0003004,000
Gemini 3.6 FlashA fast AI model for long coding sessions and iterative edits.Balanced1,40010,0001502,500
Kimi K3A Chinese AI model tuned for codingBalanced1,40010,0001502,500
GLM 5.2A perfect balance of price and capability: confidently handles analysis, code and complex tasks without the cost of flagship AI models.Balanced1,0002,000100800
DeepSeek v4 ProAn inexpensive option for code, algorithms and tasks with a clear plan.Basic5001,50050800
MiniMax M3An affordable all-purpose AI model for analysis, explanations and straightforward changes.Basic5001,50080500
Gemma 4-31The most affordable option for simple transformations, explanations and drafts.Basic17050060200
Claude Fable 5Flagship

Maximum quality for complex architecture, deep analysis and tasks where the cost of a mistake is especially high.

Input
11,000
Output
70,000
Cache read
1,300
Cache write
18,000
Claude Opus 5Flagship

For large refactorings, ambiguous requirements and thorough code review.

Input
5,200
Output
35,000
Cache read
250
Cache write
9,000
GPT 5.6 Sol ProFlagship

For the most complex reasoning, architectural decisions and mission-critical code.

Input
5,000
Output
40,000
Cache read
600
Cache write
9,000
Claude Sonnet 5Advanced

The main strong choice for development, complex logic and large edits without the flagship price tag.

Input
2,500
Output
15,000
Cache read
350
Cache write
4,000
GPT 5.6 Terra ProAdvanced

A strong trade-off between development and analysis without the top-end cost.

Input
2,500
Output
24,000
Cache read
350
Cache write
4,000
Gemini 3.1 ProAdvanced

For multi-step analysis, planning and working with a large context.

Input
2,200
Output
18,000
Cache read
220
Cache write
3,500
Grok 4.5Advanced

For research tasks, weighing alternatives and unconventional solutions.

Input
2,100
Output
10,000
Cache read
300
Cache write
4,000
Gemini 3.6 FlashBalanced

A fast AI model for long coding sessions and iterative edits.

Input
1,400
Output
10,000
Cache read
150
Cache write
2,500
Kimi K3Balanced

A Chinese AI model tuned for coding

Input
1,400
Output
10,000
Cache read
150
Cache write
2,500
GLM 5.2Balanced

A perfect balance of price and capability: confidently handles analysis, code and complex tasks without the cost of flagship AI models.

Input
1,000
Output
2,000
Cache read
100
Cache write
800
DeepSeek v4 ProBasic

An inexpensive option for code, algorithms and tasks with a clear plan.

Input
500
Output
1,500
Cache read
50
Cache write
800
MiniMax M3Basic

An affordable all-purpose AI model for analysis, explanations and straightforward changes.

Input
500
Output
1,500
Cache read
80
Cache write
500
Gemma 4-31Basic

The most affordable option for simple transformations, explanations and drafts.

Input
170
Output
500
Cache read
60
Cache write
200

Query Console services

coins per use
  • Processor generation (.epf) 1
  • Report generation (.erf) 1

Actual usage depends on context and response size. Cached tokens help reduce the cost of long sessions.