LLM Index
Overview and comparison of available providers and models
Provider | Model | Mode | Status | Release date | Deprecation date | Max. tokens | Token price in / out (1k) | Links |
---|---|---|---|---|---|---|---|---|
AI21 | Jurassic 2 Light | completion | available | Mar 9, 2023 | 8,192 | 0.003 | ||
Jurassic 2 Mid | completion | available | Mar 9, 2023 | 8,192 | 0.01 | |||
Jurassic 2 Ultra | completion | available | Mar 9, 2023 | 8,192 | 0.015 | |||
Aleph Alpha | Luminous Base | completion | available | Apr 14, 2023 | 2,048 | 0.03 | ||
Luminous Base Control | chat | available | Apr 14, 2023 | 2,048 | 0.0375 | |||
Luminous Extended | completion | available | Apr 14, 2023 | 2,048 | 0.045 | |||
Luminous Extended Control | chat | available | Apr 14, 2023 | 2,048 | 0.05625 | |||
Luminous Supreme | completion | available | Apr 14, 2023 | 2,048 | 0.175 | |||
Luminous Supreme Control | chat | available | Apr 14, 2023 | 2,048 | 0.21875 | |||
Anthropic | Claude 2 | chat | available | Jul 11, 2023 | 100,000 | 0.008 / 0.024 | ||
Claude 2.1 | chat | available | Nov 21, 2023 | 200,000 | 0.008 / 0.024 | |||
Claude Instant 1 | chat | available | Mar 14, 2023 | 100,000 | 0.00163 / 0.00551 | |||
Claude Instant 1.2 | chat | available | Sep 21, 2023 | 100,000 | 0.00163 / 0.00551 | |||
Cohere | Command | completion | available | 4,096 | 0.015 | |||
Command Light | completion | available | 4,096 | 0.015 | ||||
Command Nightly | completion | available | 4,096 | 0.015 | ||||
Deep Infra | Code Llama 34B Instruct HF | chat | Aug 24, 2023 | 4,096 | 0.0006 | |||
Llama 2 13B Chat HF | chat | available | Jul 18, 2023 | 4,096 | 0.00035 | |||
Llama 2 70B Chat HF | chat | available | Jul 18, 2023 | 6,144 | 0.001875 | |||
Llama 2 7B Chat HF | chat | available | Jul 18, 2023 | 4,096 | 0.0002 | |||
Mistral 7B Instruct v0.1 | chat | available | Sep 27, 2023 | 4,096 | 0.0002 | |||
NLP Cloud | Chat Dolphin | chat | available | 2,048 | 0.02 | |||
Dolphin | completion | available | 2,048 | 0.02 | ||||
OpenAI | Ada 001 | completion | available | Jun 11, 2020 | Jan 4, 2024 | 2,049 | 0.0004 | |
Babbage 001 | completion | available | Jun 11, 2020 | Jan 4, 2024 | 2,049 | 0.0005 | ||
Curie 001 | completion | available | Jun 11, 2020 | Jan 4, 2024 | 2,049 | 0.002 | ||
DaVinci 003 | completion | available | Nov 28, 2022 | Jan 4, 2024 | 4,097 | 0.02 | ||
GPT-3.5 Turbo | chat | available | Nov 30, 2022 | 4,096 | 0.0015 / 0.002 | |||
GPT-3.5 Turbo 0301 | chat | available | 4,096 | 0.0015 / 0.002 | ||||
GPT-3.5 Turbo 0613 | chat | available | 4,096 | 0.0015 / 0.002 | ||||
GPT-3.5 Turbo 1106 | chat | available | Nov 6, 2023 | 16,385 | 0.001 / 0.002 | |||
GPT-3.5 Turbo 16k | chat | available | Nov 30, 2022 | 16,384 | 0.003 / 0.004 | |||
GPT-3.5 Turbo 16k 0301 | chat | available | 16,384 | 0.003 / 0.004 | ||||
GPT-3.5 Turbo 16k 0613 | chat | available | 16,384 | 0.003 / 0.004 | ||||
GPT-3.5 Turbo Instruct | completion | available | 4,096 | 0.0015 / 0.002 | ||||
GPT-4 | chat | available | Mar 14, 2023 | 8,192 | 0.03 / 0.06 | |||
GPT-4 0301 | chat | available | 8,192 | 0.03 / 0.06 | ||||
GPT-4 0613 | chat | available | 8,192 | 0.03 / 0.06 | ||||
GPT-4 32k | chat | available | Mar 14, 2023 | 32,768 | 0.06 / 0.12 | |||
GPT-4 32k 0301 | chat | available | 32,768 | 0.06 / 0.12 | ||||
GPT-4 32k 0613 | chat | available | 32,768 | 0.06 / 0.12 | ||||
GPT-4 Turbo | chat | available | Nov 6, 2023 | 128,000 | 0.01 / 0.03 | |||
PaLM 2 | Chat Bison | chat | available | May 10, 2023 | 4,096 | 0.0005 | ||
Code Bison | completion | available | May 10, 2023 | 6,144 | 0.0005 | |||
Text Bison | completion | available | May 10, 2023 | 8,192 | 0.0005 | |||
Perplexity | Code Llama 34B | chat | available | Oct 4, 2023 | 4,096 | 0 | ||
Llama 2 70B | chat | available | Oct 4, 2023 | 4,096 | 0 | |||
Mistral 7B | chat | available | Oct 4, 2023 | 4,096 | 0 | |||
OpenHermes 2 Mistral 7B | chat | available | Oct 16, 2023 | 4,096 | 0 | |||
OpenHermes 2.5 Mistral 7B | chat | available | Oct 16, 2023 | 4,096 | 0 | |||
pplx 70B chat alpha | chat | available | Oct 27, 2023 | 4,096 | 0 | |||
pplx 7B chat alpha | chat | available | Oct 27, 2023 | 4,096 | 0 | |||
xAI | Grok 1 | chat | Nov 6, 2020 | 25,000 |
You can find an overview of model parameter support for each available model in the LiteLLM docs.
If you are missing a provider or model that you would like to use inside PROMPTMETHEUS, please don't hesitate to request it. The goal is to support all relevant foundation models and eventually even custom models. Shoutout to Krrish and Ishaan from LiteLLM who make this possible.
Disclaimer: Even though we try to keep this list accurate and up-to-date, please do not rely on the presented information for critical use cases and always double check official sources.
How to choose the right LLM for my use case?
Unfortunately, there is no universal answer to this question. It depends on your use case, your budget, and your personal preferences. A good strategy to go about it is to do the following:
If you are new to the game, start with the industry standard, GPT-3.5 for simple tasks and GPT-4 for more complex ones. Use PROMPTMETHEUS to experiment until you get satisfying and reproducable results. If you do not get anywhere with OpenAI, try different providers. Once you have a working prompt, try to optimize it for performance, speed, reliability, and cost by comparing different LLMs and model parameters. As a rule of thumb, the cheapest model which is fast enough and does the job is the best one.
If you are more experienced with Prompt Engineering and AI Development, start with the model that historically worked best for use cases similar to the one you have at hand. But before you go into fine-tuning of your prompt, take an early version and execute it with a few different LLMs to see which one is most promising. Use that one to fine-tune your prompt. Once you achieve great results, revisit your model choice and optimize for performance, speed, and reliability.
Please also check the "Building a Prompt Engineering IDE" and "Prompt Engineering Tips & Tricks" posts for more information.