MythoMax 13B

Text input Text output
Author's Description

One of the highest performing and most popular fine-tunes of Llama 2 13B, with rich descriptions and roleplay. #merge

Key Specifications
Cost
$$
Context
4K
Parameters
13B
Released
Jul 01, 2023
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Min P Max Tokens Presence Penalty Seed Frequency Penalty Top P Temperature Response Format Stop
Features

This model supports the following features:

Response Format
Performance Summary

MythoMax 13B, a fine-tuned Llama 2 13B model from gryphe, demonstrates exceptional speed and cost efficiency. It consistently ranks among the fastest models and offers highly competitive pricing across all benchmarks. While its overall reliability is not explicitly detailed, the model's consistent performance in providing responses across various tests suggests a stable operational profile. Performance across specific benchmarks reveals significant variability. MythoMax 13B excels in Email Classification, achieving 86.0% accuracy, indicating a strong capability in categorizing structured text based on context. However, its performance in other areas is notably low. In Coding, Instruction Following, Reasoning, Ethics, and General Knowledge, accuracy scores are very low, often in the single digits or even 0%. This suggests that while the model is highly efficient in terms of speed and cost, its current capabilities for complex tasks requiring deep understanding, logical inference, ethical judgment, or broad factual recall are limited. Its strengths lie in specific classification tasks, while areas like coding, reasoning, and general knowledge represent significant weaknesses.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.06
Completion $0.06

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
DeepInfra
DeepInfra | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Parasail
Parasail | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Novita
Novita | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Together
Together | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Together
Together | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Mancer 2
Mancer 2 | gryphe/mythomax-l2-13b 8K $0.06 / 1M tokens $0.06 / 1M tokens
NextBit
NextBit | gryphe/mythomax-l2-13b 4K $0.06 / 1M tokens $0.06 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Free Executions Accuracy Cost Duration