Author's Description
A wild 7B parameter model that merges several models using the new task_arithmetic merge method from mergekit. List of merged models: - NousResearch/Nous-Capybara-7B-V1.9 - [HuggingFaceH4/zephyr-7b-beta](/models/huggingfaceh4/zephyr-7b-beta) - lemonilia/AshhLimaRP-Mistral-7B - Vulkane/120-Days-of-Sodom-LoRA-Mistral-7b - Undi95/Mistral-pippa-sharegpt-7b-qlora #merge #uncensored
Key Specifications
Supported Parameters
This model supports the following parameters:
Performance Summary
Toppy M 7B, a 7B parameter model from undi95, demonstrates moderate speed performance, ranking in the 29th percentile across five benchmarks. It offers competitive pricing, placing in the 45th percentile. The model exhibits notable weaknesses in accuracy across most evaluated categories. In General Knowledge, it achieved only 16.3% accuracy (12th percentile), and in Ethics, it scored 28.0% (12th percentile). Its performance in Coding was particularly low at 2.0% accuracy (8th percentile). While it showed a relatively better accuracy of 92.0% in Email Classification, this still places it in the 20th percentile. Instruction Following also presented a challenge, with 31.1% accuracy (32nd percentile) and a particularly long duration. Overall, Toppy M 7B struggles with tasks requiring deep understanding and complex reasoning, but its competitive pricing might be a consideration for specific, less accuracy-critical applications.
Model Pricing
Current Pricing
Feature | Price (per 1M tokens) |
---|---|
Prompt | $0.8 |
Completion | $1.2 |
Price History
Available Endpoints
Provider | Endpoint Name | Context Length | Pricing (Input) | Pricing (Output) |
---|---|---|---|---|
Featherless
|
Featherless | undi95/toppy-m-7b | 4K | $0.8 / 1M tokens | $1.2 / 1M tokens |
Benchmark Results
Benchmark | Category | Reasoning | Strategy | Free | Executions | Accuracy | Cost | Duration |
---|
Other Models by undi95
|
Released | Params | Context |
|
Speed | Ability | Cost |
---|---|---|---|---|---|---|---|
ReMM SLERP 13B | Jul 21, 2023 | 13B | 4K |
Text input
Text output
|
★★★ | ★ | $$$$ |