Timothyemmanuel/Arakandar

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:3.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Sep 20, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Gated Featherless Exclusive Cold

Timothyemmanuel/Arakandar is a 3.1 billion parameter causal language model developed by Timothyemmanuel, featuring a 32,768-token context window. This base model is specifically designed as a local market research assistant, intended to explain structured market evidence rather than act as an autonomous trading system. It integrates with a deterministic LightGBM signal model and purpose-built tools to provide grounded explanations of market values, news, and workflow.

Loading preview...

Arakandar 3B Base Model Overview

Arakandar is a 3.1 billion parameter causal language model developed by Timothyemmanuel, specifically engineered as a local market research assistant. Its primary function is to explain structured market evidence provided by an application, rather than generating autonomous trading decisions. The model features a substantial 32,768-token context window and utilizes grouped-query attention with 16 query heads and 2 key/value heads.

Key Capabilities

  • Market Evidence Explanation: Designed to interpret and explain supplied market signals, values, news evidence, and workflow.
  • Integration with Deterministic Signals: Works in conjunction with a LightGBM signal model, where the LightGBM remains authoritative for BUY, HOLD, or SELL decisions.
  • Grounded Responses: Explicitly constrained to avoid inventing prices, news, probabilities, or trades, focusing solely on explaining provided data.
  • Modular Specialization: The base model can be specialized for conversational tasks via a separately distributed LoRA adapter.

Good For

  • Financial Market Research: Ideal for applications requiring detailed, grounded explanations of market data and signals.
  • Assisted Decision Making: Supports human analysts by providing context and rationale for market movements and trading signals.
  • Integrating with Existing Analytical Tools: Designed to complement deterministic models like LightGBM, enhancing their output with natural language explanations.
  • Applications Requiring High Context: The 32,768-token context window is beneficial for analyzing extensive market reports or historical data.

Finetunes
1 models