Understand and compare
Gemini Pro
vs.
Gemini Ultra
Try
Podial
Turn your documents into engaging podcast discussions.
Overview
Gemini Pro
|
Gemini Ultra
|
|
|---|---|---|
|
Provider
The entity that provides this model.
|
Google
|
Google
|
|
Input Context Window
The number of tokens supported by the input context window.
|
32.8K
characters
|
32.8K
characters
|
|
Maximum Output Tokens
The number of tokens that can be generated by the model in a single request.
|
8,192
characters
|
8,192
characters
|
|
Release Date
When the model was first released.
|
2023-12-13
|
Unknown
|
|
Knowledge Cutoff
Limit on the knowledge base used by the model.
|
Unknown
|
Unknown
|
|
Open Source
|
|
|
|
API Providers
The providers that offer this model. (This is not an exhaustive list.)
|
|
|
Pricing
|
Gemini Pro
|
Gemini Ultra
|
|
|---|---|---|
|
Input
Cost of input data provided to the model.
|
Pricing not available.
|
Pricing not available.
|
|
Output
Cost of output tokens generated by the model.
|
Pricing not available.
|
Pricing not available.
|
Benchmarks
Compare relevant benchmarks between Gemini Pro
and Gemini Ultra.
|
Gemini Pro
|
Gemini Ultra
|
|
|---|---|---|
|
MMLU
Evaluating LLM knowledge acquisition in zero-shot and few-shot settings.
|
71.8
(5-shot)
|
83.7
(5-shot)
|
|
MMMU
A wide ranging multi-discipline and multimodal benchmark.
|
47.9
(pass@1)
|
59.4
(0-shot pass@1)
|
|
HellaSwag
A challenging sentence completion benchmark.
|
84.7
(10-shot)
|
Benchmark not available.
|
|
GSM8K
Grade-school math problems benchmark.
|
77.9
(11-shot)
|
88.9
(11-shot)
|
|
HumanEval
A benchmark to measure functional correctness for synthesizing programs from docstrings.
|
67.7
(0-shot)
|
74.4
(0-shot)
|
|
MATH
Benchmark performance on Math problems ranging across 5 levels of difficulty and 7 sub-disciplines.
|
32.6
(4-shot Minerva Prompt)
|
53.2
(4-shot Minerva Prompt)
|
Gemini Pro, developed by Google, features a context window of 32768 tokens. The model costs 0.0125 cents per thousand tokens for input and 0.0375 cents per thousand tokens for output. It was released on December 13, 2023, and has achieved a score of 47.9 in the MMMU benchmark with a "pass@1" caveat and a score of 71.8 in the MMLU benchmark in a 5-shot scenario.
Gemini Ultra, developed by Google, features a large context window of 32768 tokens. The model has excelled in benchmarks like MMMU with a score of 59.4 in a 0-shot pass@1 scenario and MMLU with a score of 83.7 in a 5-shot scenario.
GPT-4 Turbo 0125
Llama 3 8B Instruct
Claude 2
Measure & Improve LLM
Product Performance.
Get Started
© 2023-2024 Context