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
6,287 results
deepseek

deepseek-llm-67b-base

Models
Context window
4,096 tokens
Open weights
Yes
Model license
other
Architecture
llama
Radar · unrated0/4 core tests · not enough evidence
deepseek

deepseek-llm-7b-base

Models
Context window
4,096 tokens
Open weights
Yes
Model license
other
Architecture
llama
Radar · unrated0/4 core tests · not enough evidence
deepseek

deepseek-llm-7b-chat

Models
Context window
4,096 tokens
Open weights
Yes
Model license
other
Architecture
llama
Radar · unrated0/4 core tests · not enough evidence
deepseek

deepseek-math-7b-base

Models
Context window
4,096 tokens
Open weights
Yes
Model license
other
Architecture
llama
Radar · unrated0/4 core tests · not enough evidence
deepseek

deepseek-math-7b-instruct

Models
Context window
4,096 tokens
Open weights
Yes
Model license
other
Architecture
llama
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-Math-V2

Models
Context window
163,840 tokens
Open weights
Yes
Model license
apache-2.0
Architecture
deepseek_v32
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-R1-Zero

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v3
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V3.1

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v3
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V3.1-Base

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v3
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V3.1-Terminus

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v3
Arena 1417.5Human preference · 3,607 votes
deepseek

DeepSeek-V3.2-Exp-Base

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v32
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V3.2-Speciale

Models
Context window
163,840 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v32
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V4-Flash-DSpark

Models
Context window
1,048,576 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v4
Radar · unrated0/4 core tests · not enough evidence
deepseek

DeepSeek-V4-Pro-DSpark

Models
Context window
1,048,576 tokens
Open weights
Yes
Model license
mit
Architecture
deepseek_v4
Radar · unrated0/4 core tests · not enough evidence
microsoft

DialoGPT-large

Models
Context window
1,024 tokens
Open weights
Yes
Model license
mit
Architecture
gpt2
Radar · unrated0/4 core tests · not enough evidence
microsoft

DialoGPT-medium

Models
Context window
1,024 tokens
Open weights
Yes
Model license
mit
Architecture
gpt2
Radar · unrated0/4 core tests · not enough evidence
nvidia

Eagle2.5-8B

Models
Open weights
Yes
Model license
other
Checkpoint tensor payload
15.03 GiB
Radar · unrated0/4 core tests · not enough evidence
nvidia

EGM-4B-SFT

Models
Context window
262,144 tokens
Open weights
Yes
Model license
apache-2.0
Architecture
qwen3_vl
Radar · unrated0/4 core tests · not enough evidence
nvidia

EGM-8B-SFT

Models
Context window
262,144 tokens
Open weights
Yes
Model license
apache-2.0
Architecture
qwen3_vl
Radar · unrated0/4 core tests · not enough evidence
Unspecified

elevenlabs/scribe_v1

Models*
API mode
Audio transcription*
Radar · unrated0/4 core tests · not enough evidence
Unspecified

embed-english-light-v3.0

Models*
Maximum input
512 tokens*
API mode
Embeddings*
Radar · unrated0/4 core tests · not enough evidence

* Provisional data