What are the differences?

Between the OpenChat 3.5 0106 and Gemini Ultra LLM models, which follows best instructions?

Compare

OpenChat logo

to

Google logo
OpenChat logo

OpenChat 3.5 0106

OpenChat

Google logo

Gemini Ultra

Google

Overview

OpenChat logo

OpenChat 3.5 0106

Google logo

Gemini Ultra

Provider

Organization responsible for this model.

OpenChat logo

OpenChat

Google logo

Google

Input Context Window

The total number of tokens that the input context window can accommodate.

8.2K
33K

Maximum Output Tokens

The maximum number of tokens this model can produce in one operation.

4.1K
8.2K

Release Date

The initial release date of the model.

November 1, 2023

24 months ago

December 6, 2023

23 months ago

Knowledge Cutoff

The latest date for which the information provided is considered reliable and current.

2024/1

2023/11

Pricing

OpenChat logo

OpenChat 3.5 0106

Google logo

Gemini Ultra

Input

Costs associated with the data input to the model.

Not specified.

Not specified.

Output

Costs associated with the tokens produced by the model.

Not specified.

Not specified.

Benchmark

OpenChat logo

OpenChat 3.5 0106

Google logo

Gemini Ultra

MMLU

Assesses LLMs' ability to acquire knowledge in zero-shot and few-shot scenarios.

65.8
90

MMMU

Comprehensive benchmark covering multiple disciplines and modalities.

59.4

HellaSwag

A demanding benchmark for sentence completion tasks.

Arena Elo

Ranking metric for LMSYS Chatbot Arena.

1091

5000+ teams use Lunary to build reliable AI applications

IslandsbankiZurichNetomiCloseDHL

Building an AI chatbot?

Open-source GenAI monitoring, prompt management, and magic.

Open Source

Self Hostable

1-line Integration

Prompt Templates

Chat Replays

Analytics

Topic Classification

Agent Tracing

Custom Dashboards

Score LLM responses

PII Masking

Feedback Tracking

Open Source

Self Hostable

1-line Integration

Prompt Templates

Chat Replays

Analytics

Topic Classification

Agent Tracing

Custom Dashboards

Score LLM responses

PII Masking

Feedback Tracking