Understand and compare
GPT-4o 2024-08-06
vs.
Gemini Flash
Try
Podial
Turn your documents into engaging podcast discussions.
Overview
GPT-4o 2024-08-06 was released
3 months after
Gemini Flash.
GPT-4o 2024-08-06
|
Gemini Flash
|
|
|---|---|---|
|
Provider
The entity that provides this model.
|
OpenAI
|
Google
|
|
Input Context Window
The number of tokens supported by the input context window.
|
128K
tokens
|
1M
tokens
|
|
Maximum Output Tokens
The number of tokens that can be generated by the model in a single request.
|
16.4K
tokens
|
8,192
tokens
|
|
Release Date
When the model was first released.
|
2024-08-06
|
2024-05-14
|
|
Knowledge Cutoff
Limit on the knowledge base used by the model.
|
October 2023
|
November 2023
|
|
Open Source
|
|
|
|
API Providers
The providers that offer this model. (This is not an exhaustive list.)
|
|
|
Pricing
GPT-4o 2024-08-06 is
roughly 20x more expensive compared
to Gemini Flash for input tokens and
roughly 27x more expensive
for output tokens.
|
GPT-4o 2024-08-06
|
Gemini Flash
|
|
|---|---|---|
|
Input
Cost of input data provided to the model.
|
$2.50
per million tokens
|
$0.13
per million tokens
|
|
Output
Cost of output tokens generated by the model.
|
$10.00
per million tokens
|
$0.38
per million tokens
|
Benchmarks
Compare relevant benchmarks between GPT-4o 2024-08-06
and Gemini Flash.
|
GPT-4o 2024-08-06
|
Gemini Flash
|
|
|---|---|---|
|
MMLU
Evaluating LLM knowledge acquisition in zero-shot and few-shot settings.
|
Benchmark not available.
|
78.9
(5-shot)
|
|
MMMU
A wide ranging multi-discipline and multimodal benchmark.
|
Benchmark not available.
|
56.1
(0-shot pass@1)
|
|
HellaSwag
A challenging sentence completion benchmark.
|
Benchmark not available.
|
86.5
(10-shot)
|
|
GSM8K
Grade-school math problems benchmark.
|
Benchmark not available.
|
86.2
(11-shot)
|
|
HumanEval
A benchmark to measure functional correctness for synthesizing programs from docstrings.
|
Benchmark not available.
|
74.3
(0-shot)
|
|
MATH
Benchmark performance on Math problems ranging across 5 levels of difficulty and 7 sub-disciplines.
|
Benchmark not available.
|
54.9
(4-shot Minerva Prompt)
|
Claude 2.1
Mistral 8x7B Instruct
Measure & Improve LLM
Product Performance.
Get Started
© 2023-2024 Context