geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk

COMPARE.

AI
Filters

Filter AI models

Brand
Benchmark coverage
Clear all filters
Type
6,274 results
vispark

Vision Large

Models
Context window
1,000,000 tokens
Maximum output
65,536 tokens
Open weights
No
Reasoning
Yes
Radar · unrated0/4 core tests · not enough evidence
vispark

Vision Medium

Models
Context window
1,000,000 tokens
Maximum output
65,536 tokens
Open weights
No
Reasoning
Yes
Radar · unrated0/4 core tests · not enough evidence
vispark

Vision Small

Models
Context window
1,000,000 tokens
Maximum output
65,536 tokens
Open weights
No
Reasoning
Yes
Radar · unrated0/4 core tests · not enough evidence
openai

GPT-4o (2024-05-13)

Models
Context window
128,000 tokens
Maximum output
4,096 tokens
Open weights
No
Reasoning
No
Arena 1300.4Human preference · 112,881 votes
openai

GPT-4o

Models
Context window
128,000 tokens
Maximum output
16,384 tokens
Open weights
No
Maximum input
128,000 tokens*
Radar · unrated0/4 core tests · not enough evidence
openai

GPT-4o (May 2024)

Models
Open weights
No*
ECI 128.2*General capability index
tii

Falcon 2 11B

Models
Open weights
Yes*
ECI 99.7*General capability index
nousresearch

Hermes 2 Pro Llama 3 8B

Models
Context window
8,192 tokens
Maximum output
8,192 tokens
Total parameters
8.03 B
Open weights
Yes
Radar · unrated0/4 core tests · not enough evidence
microsoft

phi-3-medium 14B

Models
Open weights
Yes*
ECI 114.6*General capability index
microsoft

phi-3-mini 3.8B

Models
Open weights
Yes*
ECI 110.8*General capability index
microsoft

phi-3-small 7.4B

Models
Open weights
Yes*
ECI 116.6*General capability index
meta

Llama 3-70B

Models
Open weights
Yes*
ECI 121.3*General capability index
meta

Llama 3-8B

Models
Open weights
Yes*
ECI 109.6*General capability index
mistral

Mixtral 8x22B Instruct

Models
Context window
65,536 tokens
Maximum output
52,428 tokens
Open weights
Yes
Reasoning
No
Radar · unrated0/4 core tests · not enough evidence
mistral

Mixtral 8x22B

Models
Open weights
Yes*
ECI 118.9*General capability index
microsoft

WizardLM 2 8x22B

Models
Context window
65,536 tokens
Maximum output
8,000 tokens
Open weights
Yes
Reasoning
No
Radar · unrated1/4 core tests · not enough evidenceWizardLM-2-8x22B
alibaba

Qwen-VL Max

Models
Context window
131,072 tokens
Maximum output
8,192 tokens
Open weights
No
Reasoning
No
Radar · unrated0/4 core tests · not enough evidence
alibaba

Qwen Max

Models
Context window
32,768 tokens
Maximum output
8,192 tokens
Open weights
No
Reasoning
No
Radar · unrated0/4 core tests · not enough evidence
baai

BGE Reranker V2 M3

Models
Context window
8,192 tokens
Open weights
Yes
Reasoning
No
Tool calling
No
Radar · unrated0/4 core tests · not enough evidence
anthropic

Claude Haiku 3

Models
Context window
200,000 tokens
Maximum output
4,096 tokens
Open weights
No
Reasoning
No
Arena 1194.6Human preference · 117,701 votes
anthropic

Claude 3 Haiku

Models
Open weights
No*
ECI 117.8*General capability index

* Provisional data