AionLabs: Aion-RP 1.0 (8B)

Text input Text output
Author's Description

Aion-RP-Llama-3.1-8B ranks the highest in the character evaluation portion of the RPBench-Auto benchmark, a roleplaying-specific variant of Arena-Hard-Auto, where LLMs evaluate each other’s responses. It is a fine-tuned base model...

Key Specifications
Cost
$$$
Context
32K
Parameters
8B
Released
Feb 04, 2025
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Temperature Max Tokens Top P
Performance Summary

AionLabs: Aion-RP 1.0 (8B) demonstrates exceptional performance in terms of speed and cost-efficiency, consistently ranking among the fastest models and offering highly competitive pricing across various benchmarks. This model is specifically fine-tuned for roleplaying, excelling in character evaluation within the RPBench-Auto benchmark. However, its general performance across standard benchmarks reveals significant limitations. The model exhibits a high propensity for hallucination, with only 2.0% accuracy in the Hallucinations (Baseline) test, indicating a struggle to acknowledge uncertainty. Its performance in General Knowledge (58.5% accuracy), Ethics (14.0% accuracy), Email Classification (34.0% accuracy), and Reasoning (12.0% accuracy) is notably low, placing it in the lower percentiles for these categories. A critical weakness is its complete failure in Instruction Following, achieving 0.0% accuracy. While its Coding performance is moderate at 47.0% accuracy, it still ranks in the lower quartile. The model's strength lies in its specialized roleplaying capabilities and its economic operational profile, but it struggles significantly with factual accuracy, ethical reasoning, and general instruction adherence.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.8
Completion $1.6

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
AionLabs
AionLabs | aion-labs/aion-rp-llama-3.1-8b 32K $0.8 / 1M tokens $1.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by aion-labs