TheDrummer: Anubis 70B V1.1

Text input Text output Unavailable
Author's Description

TheDrummer's Anubis v1.1 is an unaligned, creative Llama 3.3 70B model focused on providing character-driven roleplay & stories. It excels at gritty, visceral prose, unique character adherence, and coherent narratives, while maintaining the instruction following Llama 3.3 70B is known for.

Key Specifications
Cost
$$$
Context
131K
Parameters
70B
Released
Jun 29, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

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

TheDrummer's Anubis 70B V1.1 demonstrates competitive response times, performing among the faster models with a 43rd percentile speed ranking. It also offers competitive pricing, ranking in the 57th percentile. Notably, the model exhibits strong reliability with an 88% success rate, consistently providing usable responses. In terms of performance across benchmarks, Anubis 70B V1.1 shows particular strength in Email Classification, achieving perfect 100% accuracy, making it the most accurate model at its price point and among models of similar speed. It also performs well in Hallucinations (Baseline) with 96.0% accuracy, indicating a good ability to acknowledge uncertainty. Instruction Following is another area of competence, scoring 62.0% accuracy (68th percentile). However, the model exhibits notable weaknesses in Mathematics and Ethics. Its Mathematics (Baseline) score is 36.4% (19th percentile), suggesting significant challenges with complex mathematical problems. Similarly, its Ethics (Baseline) score of 89.0% (23rd percentile) indicates room for improvement in moral reasoning and ethical decision-making. General Knowledge (97.0%) and Coding (70.0%) show average performance, while Reasoning (40.0%) is also an area for development. Overall, Anubis 70B V1.1 excels in specific classification and instruction-following tasks, while mathematical and ethical reasoning present opportunities for enhancement.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.75
Completion $1

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
Parasail
Parasail | thedrummer/anubis-70b-v1.1 131K $0.75 / 1M tokens $1 / 1M tokens
NextBit
NextBit | thedrummer/anubis-70b-v1.1 4K $0.75 / 1M tokens $1 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by thedrummer