model_registry.bin
matexai_ossysplatform/models/
Registry / Model_Nodes
MODEL_DIRECTORY LATEST_SYNC: 2026-03-01
v2.0-LATEST
MATEXAI_CORE
GENERAL_PURPOSE
Premier reasoning model with enhanced multi-modal awareness and adaptive emotional intelligence.
Inference EngineGemini-1.5-Pro
Context
2M Context
Latency
240ms
Strength
Complex Logic
API_IDENTIFIER
"matexai"v3.1-STABLE
MATEXCODEX_PRO
PROGRAMMING_EXPERT
State-of-the-art coding assistant specifically optimized for system architecture and debugging.
Inference EngineDeepSeek-V3
Context
128k Context
Latency
310ms
Strength
Architecture
API_IDENTIFIER
"matexcodex"v1.0-FAST
MATEXCODEX_LITE
QUICK_SCRIPTING
Sub-millisecond latency model for rapid prototyping and small utility scripting tasks.
Inference EngineLLama-3-8B
Context
8k Context
Latency
15ms
Strength
Speed
API_IDENTIFIER
"matexcodexlite"v1.4-ORCH
MATEX_ELITE_CLUSTER
MULTI_AGENT
Self-correcting multi-agent swarm that validates responses across 5 distinct expert nodes.
Inference EngineBuildEx-Orchestrator
Context
64k Context
Latency
2.4s
Strength
Zero-Error
API_IDENTIFIER
"matexelite" SECURITY_CLEARANCE: LEVEL_4_ORCHESTRATOR
All models listed above are dynamically routed based on system load and request complexity. Pricing tiers apply to \`matexelite\` clusters. Default token weighting is normalized to Gemini-Pro equivalent for standard billing consistency.