TheDrummer: Anubis Pro 105B V1

Text input Text output
Author's Description

Anubis Pro 105B v1 is an expanded and refined variant of Meta’s Llama 3.3 70B, featuring 50% additional layers and further fine-tuning to leverage its increased capacity. Designed for advanced narrative, roleplay, and instructional tasks, it demonstrates enhanced emotional intelligence, creativity, nuanced character portrayal, and superior prompt adherence compared to smaller models. Its larger parameter count allows for deeper contextual understanding and extended reasoning capabilities, optimized for engaging, intelligent, and coherent interactions.

Key Specifications
Cost
$$$$
Context
131K
Parameters
105B
Released
Mar 10, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Presence Penalty Stop Top P Temperature Min P Seed Frequency Penalty Max Tokens
Performance Summary

TheDrummer's Anubis Pro 105B V1 demonstrates competitive performance in terms of both speed and pricing, ranking in the 45th and 41st percentiles respectively across six benchmarks. This indicates it offers a balanced proposition in the market. While specific reliability data was not provided, the model's consistent completion of benchmarks suggests a generally stable operation. Anubis Pro 105B V1 exhibits notable strengths in classification tasks, achieving perfect accuracy in Email Classification, making it the most accurate model at its price point and among models of similar speed. It also performs well in Instruction Following and Reasoning, showing above-average accuracy. However, the model struggles significantly in Coding, Ethics, and General Knowledge, where its accuracy falls into the bottom 20th percentile. Its performance in these areas, particularly Coding and General Knowledge, suggests limitations in broad factual recall and complex programming understanding despite its large parameter count. The model's description highlights its design for advanced narrative and roleplay, which aligns with its strong performance in tasks requiring nuanced understanding and adherence to instructions, but its current benchmark results do not fully reflect superior general knowledge or ethical reasoning.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.5
Completion $1

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Parasail
Parasail | thedrummer/anubis-pro-105b-v1 131K $0.5 / 1M tokens $1 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration
Other Models by thedrummer