Skip to main content

Engine Overview

Benchmarked against: Anthropic — Models overview Tools: list_models, call_model, search_web Decision: Engine Selection Guide (Captain decision, 2026-02-27)

SuperPortia operates a multi-engine fleet — six AI providers with different cost profiles, capabilities, and specializations. Unlike single-model systems, SuperPortia matches the right engine to each task, optimizing for the minimum cost that gets the job done right.


The engine fleet

Primary engine: Claude

ModelIDBest for
Opus 4.6claude-opus-4-6Architecture, decisions, complex reasoning, code operations
Sonnet 4.6claude-sonnet-4-6Standard coding, execution, analysis
Haiku 4.5claude-haiku-4-5Fast responses, high-volume tasks

Claude is the only engine with full tool access (file operations, code editing, shell commands). All other engines are text-only — they can analyze and respond but cannot modify files or run commands.

Low-cost engines

EngineProviderDefault ModelCostStrengths
GroqGroqLlama 3.3 70BFreeFast inference, research, simple analysis
GeminiGoogle2.5 Flash~$0.014/searchAuthoritative search with citations, embeddings
DeepSeekDeepSeekR1 / V3Cents/requestDeep reasoning, analysis tasks
MistralMistralLatestCents/requestEuropean alternative, general tasks
ZhipuZhipu AIGLM-5Cents/requestChinese NLP, agent tool-calling

Search-augmented engines

Two engines support built-in web search:

EngineModeUse case
Groq Search (groq-search)Groq Compound with web searchFree intel/research tasks
Gemini Search (gemini-search)Gemini + Google Search GroundingAuthoritative research with citation links

Engine capabilities matrix

CapabilityClaudeGroqGeminiDeepSeekMistralZhipu
Text generationYesYesYesYesYesYes
Web searchVia toolsCompoundGroundingNoNoNo
File operationsYesNoNoNoNoNo
Code executionYesNoNoNoNoNo
Tool callingFull MCPXML formatJSONLimitedLimitedJSON
EmbeddingsNoNoYes (embedding-001)NoNoNo
Chinese NLPGoodLimitedGoodGoodLimitedExcellent

Architecture


Checking available engines

# List all engines, models, and API key status
list_models()

Returns available providers, their default models, and whether API keys are configured.


Known limitations

EngineLimitationWorkaround
GroqXML tool calling format — breaks nested agent modeUse flat mode (automatic)
GroqToken explosion on ingest_text callsIngest handled at orchestrator layer
GroqHallucinated version numbers for tech infoVerify with Pre-Flight Check
DeepSeekLimited tool calling supportUse flat mode
MistralLimited tool calling supportUse flat mode
All non-ClaudeNo file/shell operationsOnly Claude engine for WOs requiring file ops

PageRelationship
Choosing an EngineDecision framework
PricingCost details per engine
Usage & CostTracking and optimization
Dispatch EngineHow engines are invoked