LLM Database

Categories

Year:
Showing 120 LLMs

Aya 23-35B

MULTILINGUAL
Cohere2024Multilingual Transformer
Parameters
35B
Context
8K
Speed
20-40 tok/s
Pricing
Free
Text
MMLU: 71.4% • HumanEval: 31.7%

BERT Base Uncased

LANGUAGE UNDERSTANDING
Google2018Bidirectional Encoder Transformer
Parameters
110M
Context
512
Speed
200-400 tok/s
Pricing
Free
Text
MMLU: 62.2% • HumanEval: 0%

BERT Large Uncased

LANGUAGE UNDERSTANDING
Google2018Bidirectional Encoder Transformer
Parameters
340M
Context
512
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 68.1% • HumanEval: 0%

BLOOM 176B

MULTILINGUAL
BigScience2022Multilingual Transformer Decoder
Parameters
176B
Context
2K
Speed
8-20 tok/s
Pricing
Free
Text
MMLU: 55.1% • HumanEval: 18.1%

BLOOMZ 176B

MULTILINGUAL
BigScience2022Instruction-tuned Multilingual Transformer
Parameters
176B
Context
2K
Speed
8-20 tok/s
Pricing
Free
Text
MMLU: 57.8% • HumanEval: 22.7%

Claude 1.0

GENERAL PURPOSE
Anthropic2022Constitutional AI Transformer
Parameters
52B
Context
9K
Speed
25-50 tok/s
Pricing
$0.01/1K
Text
MMLU: 75% • HumanEval: 71.2%

Claude 1.2

GENERAL PURPOSE
Anthropic2022Constitutional AI Transformer
Parameters
52B
Context
9K
Speed
25-50 tok/s
Pricing
$0.01/1K
Text
MMLU: 76.2% • HumanEval: 72.5%

Claude 1.3

GENERAL PURPOSE
Anthropic2022Constitutional AI Transformer
Parameters
52B
Context
9K
Speed
25-50 tok/s
Pricing
$0.01/1K
Text
MMLU: 77.5% • HumanEval: 73.8%

Claude 2.0

GENERAL PURPOSE
Anthropic2023Constitutional AI Transformer
Parameters
Unknown
Context
100K
Speed
20-40 tok/s
Pricing
$0.008/1K
Text
MMLU: 78.5% • HumanEval: 71.2%

Claude 2.1

GENERAL PURPOSE
Anthropic2023Constitutional AI Transformer
Parameters
Unknown
Context
200K
Speed
20-40 tok/s
Pricing
$0.008/1K
Text
MMLU: 81.2% • HumanEval: 75.5%

Claude 3 Haiku

SMALL EFFICIENT
Anthropic2024Transformer
Parameters
Unknown
Context
200K
Speed
80-150 tok/s
Pricing
$0.00025/1K
TextVision
MMLU: 75.2% • HumanEval: 75.9%

Claude 3 Opus

GENERAL PURPOSE
Anthropic2024Transformer
Parameters
Unknown
Context
200K
Speed
15-35 tok/s
Pricing
$0.015/1K
TextVision
MMLU: 86.8% • HumanEval: 84.9%

Claude 3 Sonnet

GENERAL PURPOSE
Anthropic2024Transformer
Parameters
Unknown
Context
200K
Speed
30-70 tok/s
Pricing
$0.003/1K
TextVision
MMLU: 79% • HumanEval: 73%

Claude 3.5 Haiku

SMALL EFFICIENT
Anthropic2024Transformer
Parameters
Unknown
Context
200K
Speed
100-200 tok/s
Pricing
$0.0008/1K
TextVision
MMLU: 82.2% • HumanEval: 85.9%

Claude 3.5 Sonnet

GENERAL PURPOSE
Anthropic2024Transformer
Parameters
~200B
Context
200K
Speed
30-70 tok/s
Pricing
$0.003/1K
TextVision
MMLU: 88.3% • HumanEval: 92%

Claude 3.5 Sonnet (Updated)

GENERAL PURPOSE
Anthropic2024Transformer
Parameters
~200B
Context
200K
Speed
30-70 tok/s
Pricing
$0.003/1K
TextVision
MMLU: 88.7% • HumanEval: 92%

Claude 4 Opus

REASONING
Anthropic2025Constitutional AI Transformer
Parameters
~400B
Context
1M
Speed
15-35 tok/s
Pricing
$0.015/1K
TextVisionDocument
MMLU: 92.3% • HumanEval: 94.7%

Claude 4 Sonnet

GENERAL PURPOSE
Anthropic2025Constitutional AI Transformer
Parameters
~200B
Context
500K
Speed
30-70 tok/s
Pricing
$0.003/1K
TextVisionDocument
MMLU: 89.7% • HumanEval: 92.8%

Code Llama-34B Instruct

CODE SPECIALIZED
Meta2023Decoder-only Transformer
Parameters
34B
Context
16K
Speed
20-40 tok/s
Pricing
Free
Text
MMLU: 56.8% • HumanEval: 48.8%

CodeQwen1.5 7B

CODE SPECIALIZED
Alibaba2024Code-specialized Transformer
Parameters
7B
Context
64K
Speed
60-120 tok/s
Pricing
Free
Text
MMLU: 62.1% • HumanEval: 83.5%

Codestral 22B v0.1

CODE SPECIALIZED
Mistral AI2024Code-specialized Transformer
Parameters
22B
Context
32K
Speed
30-60 tok/s
Pricing
$0.001/1K
Text
MMLU: 67.1% • HumanEval: 81.4%

Codestral Latest

CODE SPECIALIZED
Mistral AI2024Enhanced Code-specialized Transformer
Parameters
22B
Context
32K
Speed
35-70 tok/s
Pricing
$0.001/1K
Text
MMLU: 69.4% • HumanEval: 84.2%

Codex

CODE SPECIALIZED
OpenAI2021Transformer Decoder
Parameters
12B
Context
8K
Speed
30-70 tok/s
Pricing
$0.02/1K
Text
MMLU: 35% • HumanEval: 47%

Command

GENERAL PURPOSE
Cohere2022Transformer Decoder
Parameters
52B
Context
4K
Speed
20-40 tok/s
Pricing
$0.0015/1K
Text
MMLU: 68.3% • HumanEval: 25.7%

Command R

RAG SPECIALIZED
Cohere2024Retrieval-Augmented Transformer
Parameters
35B
Context
128K
Speed
25-50 tok/s
Pricing
$0.0005/1K
Text
MMLU: 73.8% • HumanEval: 40.7%

Command R+

RAG SPECIALIZED
Cohere2024Large Retrieval-Augmented Transformer
Parameters
104B
Context
128K
Speed
10-25 tok/s
Pricing
$0.0025/1K
Text
MMLU: 80.2% • HumanEval: 56.1%

DeepSeek-Coder 1.3B Instruct

CODE SPECIALIZED
DeepSeek2023Code-specialized Transformer
Parameters
1.3B
Context
16K
Speed
200-400 tok/s
Pricing
Free
Text
MMLU: 37.8% • HumanEval: 65.8%

DeepSeek-Coder 33B Instruct

CODE SPECIALIZED
DeepSeek2023Large Code-specialized Transformer
Parameters
33B
Context
16K
Speed
20-40 tok/s
Pricing
Free
Text
MMLU: 58.8% • HumanEval: 78.6%

DeepSeek-Coder-V2 236B

CODE SPECIALIZED
DeepSeek2024Mixture of Experts Code Transformer
Parameters
236B
Context
128K
Speed
8-20 tok/s
Pricing
$0.00027/1K
Text
MMLU: 75.9% • HumanEval: 90.2%

DeepSeek-Coder-V2 Lite Instruct

CODE SPECIALIZED
DeepSeek2024Enhanced Code Transformer
Parameters
16B
Context
128K
Speed
40-80 tok/s
Pricing
$0.00014/1K
Text
MMLU: 60.1% • HumanEval: 81.1%

DeepSeek-LLM 67B Chat

GENERAL PURPOSE
DeepSeek2023Instruction-tuned Transformer
Parameters
67B
Context
4K
Speed
12-25 tok/s
Pricing
Free
Text
MMLU: 71.3% • HumanEval: 37.6%

DeepSeek-LLM 7B Base

GENERAL PURPOSE
DeepSeek2023Transformer Decoder
Parameters
7B
Context
4K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 48.2% • HumanEval: 26.6%

DeepSeek-Math 7B Base

REASONING SPECIALIZED
DeepSeek2024Math-specialized Transformer
Parameters
7B
Context
4K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 64.7% • HumanEval: 43.6%

DeepSeek-Math 7B Instruct

REASONING SPECIALIZED
DeepSeek2024Instruction-tuned Math Transformer
Parameters
7B
Context
4K
Speed
75-140 tok/s
Pricing
Free
Text
MMLU: 67.2% • HumanEval: 45.1%

DeepSeek-V2 Chat

GENERAL PURPOSE
DeepSeek2024Mixture of Experts Transformer
Parameters
236B
Context
128K
Speed
10-25 tok/s
Pricing
$0.00014/1K
Text
MMLU: 78.5% • HumanEval: 89.6%

DeepSeek-V3 671B

GENERAL PURPOSE
DeepSeek2024Large Mixture of Experts Transformer
Parameters
671B
Context
128K
Speed
3-10 tok/s
Pricing
$0.00027/1K
Text
MMLU: 88.5% • HumanEval: 92.2%

DialoGPT Large

CONVERSATIONAL
Microsoft2019GPT-2 based Dialogue Transformer
Parameters
762M
Context
1K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 35.7% • HumanEval: 0%

DialoGPT Small

CONVERSATIONAL
Microsoft2019GPT-2 based Dialogue Transformer
Parameters
117M
Context
1K
Speed
300-500 tok/s
Pricing
Free
Text
MMLU: 28.3% • HumanEval: 0%

FLAN-T5 XL

INSTRUCTION FOLLOWING
Google2022Instruction-tuned Text-to-Text Transformer
Parameters
3B
Context
512
Speed
40-80 tok/s
Pricing
Free
Text
MMLU: 52.4% • HumanEval: 22%

FLAN-T5 XXL

INSTRUCTION FOLLOWING
Google2022Instruction-tuned Text-to-Text Transformer
Parameters
11B
Context
512
Speed
15-30 tok/s
Pricing
Free
Text
MMLU: 55.1% • HumanEval: 30.2%

Falcon 180B Chat

GENERAL PURPOSE
Technology Innovation Institute2023RefinedWeb-trained Transformer
Parameters
180B
Context
2K
Speed
8-20 tok/s
Pricing
Free
Text
MMLU: 70.4% • HumanEval: 35%

GPT-1

FOUNDATIONAL
OpenAI2018Transformer Decoder
Parameters
117M
Context
512
Speed
100-300 tok/s
Pricing
Research only
Text
MMLU: 20% • HumanEval: 0%

GPT-2 Large

FOUNDATIONAL
OpenAI2019Transformer Decoder
Parameters
774M
Context
1K
Speed
40-100 tok/s
Pricing
Open Source
Text
MMLU: 29% • HumanEval: 1.8%

GPT-2 Medium

FOUNDATIONAL
OpenAI2019Transformer Decoder
Parameters
355M
Context
1K
Speed
60-150 tok/s
Pricing
Open Source
Text
MMLU: 28% • HumanEval: 1.5%

GPT-2 Small

FOUNDATIONAL
OpenAI2019Transformer Decoder
Parameters
124M
Context
1K
Speed
80-200 tok/s
Pricing
Open Source
Text
MMLU: 25% • HumanEval: 1%

GPT-2 XL

FOUNDATIONAL
OpenAI2019Transformer Decoder
Parameters
1.5B
Context
1K
Speed
30-80 tok/s
Pricing
Open Source
Text
MMLU: 30% • HumanEval: 2%

GPT-3 Ada

FOUNDATIONAL
OpenAI2020Transformer Decoder
Parameters
350M
Context
2K
Speed
200-500 tok/s
Pricing
$0.0004/1K
Text
MMLU: 25% • HumanEval: 0%

GPT-3 Babbage

FOUNDATIONAL
OpenAI2020Transformer Decoder
Parameters
1.3B
Context
2K
Speed
100-250 tok/s
Pricing
$0.0005/1K
Text
MMLU: 30% • HumanEval: 0%

GPT-3 Curie

FOUNDATIONAL
OpenAI2020Transformer Decoder
Parameters
6.7B
Context
2K
Speed
60-150 tok/s
Pricing
$0.002/1K
Text
MMLU: 35% • HumanEval: 0%

GPT-3 Davinci

FOUNDATIONAL
OpenAI2020Transformer Decoder
Parameters
175B
Context
4K
Speed
15-40 tok/s
Pricing
$0.02/1K
Text
MMLU: 43.9% • HumanEval: 0%

GPT-3.5 Turbo

GENERAL PURPOSE
OpenAI2022Transformer Decoder
Parameters
175B
Context
16K
Speed
40-80 tok/s
Pricing
$0.0005/1K
Text
MMLU: 70% • HumanEval: 48.1%

GPT-3.5 Turbo 16K

GENERAL PURPOSE
OpenAI2023Transformer Decoder
Parameters
175B
Context
16K
Speed
35-70 tok/s
Pricing
$0.003/1K
Text
MMLU: 70% • HumanEval: 48.1%

GPT-4

MULTIMODAL
OpenAI2023Multimodal Transformer
Parameters
~1.7T
Context
8K
Speed
15-30 tok/s
Pricing
$0.03/1K
TextVision
MMLU: 86.4% • HumanEval: 67%

GPT-4 Turbo

GENERAL PURPOSE
OpenAI2024Transformer
Parameters
~1.7T
Context
128K
Speed
20-50 tok/s
Pricing
$0.01/1K
TextVision
MMLU: 86.4% • HumanEval: 67%

GPT-4-32K

MULTIMODAL
OpenAI2023Multimodal Transformer
Parameters
~1.7T
Context
32K
Speed
10-25 tok/s
Pricing
$0.06/1K
TextVision
MMLU: 86.4% • HumanEval: 67%

GPT-4.1

GENERAL PURPOSE
OpenAI2025Multimodal Transformer
Parameters
Unknown
Context
1M
Speed
25-60 tok/s
Pricing
$0.002/1K
TextVision
MMLU: 88.9% • HumanEval: 89.2%

GPT-4.1 Mini

SMALL EFFICIENT
OpenAI2025Multimodal Transformer
Parameters
Unknown
Context
1M
Speed
50-120 tok/s
Pricing
$0.0004/1K
TextVision
MMLU: 86.5% • HumanEval: 85.7%

GPT-4.1 Nano

SMALL EFFICIENT
OpenAI2025Multimodal Transformer
Parameters
Unknown
Context
1M
Speed
100-250 tok/s
Pricing
$0.0001/1K
TextVision
MMLU: 80.1% • HumanEval: 75.4%

GPT-4o

MULTIMODAL
OpenAI2024Multimodal Transformer
Parameters
~200B
Context
128K
Speed
60-120 tok/s
Pricing
$0.005/1K
TextVisionAudio
MMLU: 88.7% • HumanEval: 90.2%

GPT-4o Mini

SMALL EFFICIENT
OpenAI2024Multimodal Transformer
Parameters
~8B
Context
128K
Speed
150-300 tok/s
Pricing
$0.00015/1K
TextVision
MMLU: 82% • HumanEval: 87.2%

GPT-J 6B

GENERAL PURPOSE
EleutherAI2021GPT-style Transformer Decoder
Parameters
6B
Context
2K
Speed
100-200 tok/s
Pricing
Free
Text
MMLU: 42.1% • HumanEval: 11.6%

GPT-NeoX 20B

GENERAL PURPOSE
EleutherAI2022Enhanced GPT-style Transformer
Parameters
20B
Context
2K
Speed
30-60 tok/s
Pricing
Free
Text
MMLU: 51.6% • HumanEval: 15.4%

Gemini 1.5 Flash

MULTIMODAL
Google2024Mixture of Experts Transformer
Parameters
~20B
Context
1M
Speed
100-200 tok/s
Pricing
$0.000075/1K
TextVisionAudio
MMLU: 78.9% • HumanEval: 74.2%

Gemini 1.5 Pro

MULTIMODAL
Google2024Mixture of Experts Transformer
Parameters
~175B
Context
2M
Speed
30-60 tok/s
Pricing
$0.00125/1K
TextVisionAudio
MMLU: 85.9% • HumanEval: 71.9%

Gemini 2.0 Flash

MULTIMODAL
Google2024Next-gen Multimodal Transformer
Parameters
~25B
Context
1M
Speed
80-150 tok/s
Pricing
$0.000075/1K
TextVisionAudio
MMLU: 85.8% • HumanEval: 85.4%

Gemini 2.0 Flash Thinking

REASONING SPECIALIZED
Google2024Reasoning-enhanced Multimodal Transformer
Parameters
~25B
Context
1M
Speed
40-80 tok/s
Pricing
$0.000075/1K
TextVisionAudio
MMLU: 88.7% • HumanEval: 88.9%

Gemini Pro

GENERAL PURPOSE
Google2023Multimodal Transformer
Parameters
~175B
Context
32K
Speed
40-80 tok/s
Pricing
$0.0005/1K
Text
MMLU: 83.7% • HumanEval: 67.7%

Gemini Ultra

GENERAL PURPOSE
Google2023Multimodal Transformer
Parameters
~1.5T
Context
32K
Speed
5-15 tok/s
Pricing
$0.125/1K
Text
MMLU: 90% • HumanEval: 74.4%

Granite 3.0 8B Instruct

GENERAL PURPOSE
IBM2024Enterprise Transformer
Parameters
8B
Context
128K
Speed
60-120 tok/s
Pricing
$0.0005/1K
Text
MMLU: 75.4% • HumanEval: 58.9%

Grok-2

MULTIMODAL
xAI2024Advanced Transformer
Parameters
314B
Context
131K
Speed
5-15 tok/s
Pricing
$0.002/1K
TextVision
MMLU: 86% • HumanEval: 79.2%

InstructGPT

FOUNDATIONAL
OpenAI2022Transformer Decoder
Parameters
175B
Context
4K
Speed
20-50 tok/s
Pricing
$0.02/1K
Text
MMLU: 60% • HumanEval: 26.2%

Jamba 1.5 Large

GENERAL PURPOSE
AI21 Labs2024Mamba-Transformer Hybrid
Parameters
94B
Context
256K
Speed
12-30 tok/s
Pricing
$0.002/1K
Text
MMLU: 80.4% • HumanEval: 58.1%

Jurassic-1 Jumbo

GENERAL PURPOSE
AI21 Labs2021Transformer Decoder
Parameters
178B
Context
2K
Speed
8-20 tok/s
Pricing
$0.015/1K
Text
MMLU: 64.1% • HumanEval: 23.4%

Jurassic-2 Ultra

GENERAL PURPOSE
AI21 Labs2023Enhanced Transformer Decoder
Parameters
178B
Context
8K
Speed
10-25 tok/s
Pricing
$0.015/1K
Text
MMLU: 71.2% • HumanEval: 31.9%

LLaMA-65B

RESEARCH
Meta2023Decoder-only Transformer
Parameters
65B
Context
2K
Speed
15-30 tok/s
Pricing
Free
Text
MMLU: 63.4% • HumanEval: 23.7%

LLaMA-7B

RESEARCH
Meta2023Decoder-only Transformer
Parameters
7B
Context
2K
Speed
100-200 tok/s
Pricing
Free
Text
MMLU: 35.1% • HumanEval: 10.5%

Llama 2-70B Chat

GENERAL PURPOSE
Meta2023Decoder-only Transformer
Parameters
70B
Context
4K
Speed
12-25 tok/s
Pricing
Free
Text
MMLU: 68.9% • HumanEval: 29.9%

Llama 2-7B Chat

GENERAL PURPOSE
Meta2023Decoder-only Transformer
Parameters
7B
Context
4K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 48.9% • HumanEval: 13.1%

Llama 3-70B Instruct

GENERAL PURPOSE
Meta2024Decoder-only Transformer
Parameters
70B
Context
8K
Speed
10-20 tok/s
Pricing
Free
Text
MMLU: 82% • HumanEval: 81.7%

Llama 3-8B Instruct

GENERAL PURPOSE
Meta2024Decoder-only Transformer
Parameters
8B
Context
8K
Speed
70-120 tok/s
Pricing
Free
Text
MMLU: 68.4% • HumanEval: 62.2%

Llama 3.1-405B Instruct

GENERAL PURPOSE
Meta2024Decoder-only Transformer
Parameters
405B
Context
128K
Speed
2-5 tok/s
Pricing
Free
Text
MMLU: 88.6% • HumanEval: 89%

Llama 3.2-90B Vision

MULTIMODAL
Meta2024Multimodal Transformer
Parameters
90B
Context
128K
Speed
8-15 tok/s
Pricing
Free
TextVision
MMLU: 86.3% • HumanEval: 84.7%

Mistral 7B Instruct v0.2

INSTRUCTION FOLLOWING
Mistral AI2023Instruction-tuned Transformer with Sliding Window
Parameters
7.3B
Context
32K
Speed
70-130 tok/s
Pricing
$0.00025/1K
Text
MMLU: 65.4% • HumanEval: 36.8%

Mistral 7B v0.1

GENERAL PURPOSE
Mistral AI2023Transformer Decoder with Sliding Window Attention
Parameters
7.3B
Context
8K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 60.1% • HumanEval: 30.5%

Mistral Large 2407

GENERAL PURPOSE
Mistral AI2024Large Transformer with Enhanced Reasoning
Parameters
~123B
Context
128K
Speed
8-20 tok/s
Pricing
$0.003/1K
Text
MMLU: 84% • HumanEval: 73%

Mistral Medium

GENERAL PURPOSE
Mistral AI2024Large Transformer Decoder
Parameters
~70B
Context
32K
Speed
15-30 tok/s
Pricing
$0.0027/1K
Text
MMLU: 75.3% • HumanEval: 61.4%

Mistral Small Latest

GENERAL PURPOSE
Mistral AI2024Optimized Transformer Decoder
Parameters
~22B
Context
128K
Speed
40-80 tok/s
Pricing
$0.001/1K
Text
MMLU: 72.2% • HumanEval: 58.4%

Mixtral 8x22B

GENERAL PURPOSE
Mistral AI2024Large Sparse Mixture of Experts
Parameters
176B
Context
64K
Speed
5-15 tok/s
Pricing
Free
Text
MMLU: 77.8% • HumanEval: 45.1%

Mixtral 8x22B Instruct v0.1

GENERAL PURPOSE
Mistral AI2024Instruction-tuned Large Sparse MoE
Parameters
176B
Context
64K
Speed
4-12 tok/s
Pricing
$0.002/1K
Text
MMLU: 78.9% • HumanEval: 61.4%

Mixtral 8x7B Instruct v0.1

GENERAL PURPOSE
Mistral AI2023Instruction-tuned Sparse MoE Transformer
Parameters
46.7B
Context
32K
Speed
20-40 tok/s
Pricing
$0.0007/1K
Text
MMLU: 71.4% • HumanEval: 54.8%

Mixtral 8x7B v0.1

GENERAL PURPOSE
Mistral AI2023Sparse Mixture of Experts Transformer
Parameters
46.7B
Context
32K
Speed
25-45 tok/s
Pricing
Free
Text
MMLU: 70.6% • HumanEval: 40.2%

Nemotron-4 340B Instruct

GENERAL PURPOSE
NVIDIA2024Large Instruction-tuned Transformer
Parameters
340B
Context
4K
Speed
5-15 tok/s
Pricing
$0.0016/1K
Text
MMLU: 81.8% • HumanEval: 73.2%

Nova Pro

MULTIMODAL
Amazon2024Multimodal Transformer
Parameters
~200B
Context
300K
Speed
10-25 tok/s
Pricing
$0.0008/1K
TextVisionVideo
MMLU: 82.8% • HumanEval: 68.4%

OPT-125M

RESEARCH
Meta2022Decoder-only Transformer
Parameters
125M
Context
2K
Speed
500-1000 tok/s
Pricing
Free
Text
MMLU: 25.8% • HumanEval: 12.2%

OPT-175B

RESEARCH
Meta2022Decoder-only Transformer
Parameters
175B
Context
2K
Speed
10-20 tok/s
Pricing
Free
Text
MMLU: 42.3% • HumanEval: 18.9%

PaLM 2 Text Bison

GENERAL PURPOSE
Google2023Pathways Language Model v2
Parameters
~340B
Context
8K
Speed
25-50 tok/s
Pricing
$0.0005/1K
Text
MMLU: 78.3% • HumanEval: 37.6%

PaLM 540B

RESEARCH
Google2022Pathways Language Model
Parameters
540B
Context
2K
Speed
3-8 tok/s
Pricing
N/A
Text
MMLU: 70.7% • HumanEval: 26.2%

Phi-1 1.3B

CODE SPECIALIZED
Microsoft2023Transformer Decoder
Parameters
1.3B
Context
2K
Speed
200-400 tok/s
Pricing
Free
Text
MMLU: 42.1% • HumanEval: 50.6%

Phi-2 2.7B

GENERAL PURPOSE
Microsoft2023Transformer Decoder
Parameters
2.7B
Context
2K
Speed
150-300 tok/s
Pricing
Free
Text
MMLU: 52.7% • HumanEval: 47%

Phi-3 Medium 128K

GENERAL PURPOSE
Microsoft2024Transformer Decoder
Parameters
14B
Context
128K
Speed
25-50 tok/s
Pricing
$0.001/1K
Text
MMLU: 75.3% • HumanEval: 70.4%

Phi-3 Mini 128K

GENERAL PURPOSE
Microsoft2024Transformer Decoder
Parameters
3.8B
Context
128K
Speed
80-150 tok/s
Pricing
$0.0002/1K
Text
MMLU: 69.2% • HumanEval: 61.8%

Phi-3 Mini 4K

GENERAL PURPOSE
Microsoft2024Transformer Decoder
Parameters
3.8B
Context
4K
Speed
120-250 tok/s
Pricing
$0.0001/1K
Text
MMLU: 69% • HumanEval: 61.2%

Phi-3.5 Mini Instruct

INSTRUCTION FOLLOWING
Microsoft2024Instruction-tuned Transformer
Parameters
3.8B
Context
128K
Speed
100-200 tok/s
Pricing
$0.0001/1K
Text
MMLU: 70.9% • HumanEval: 68.1%

Phi-3.5 MoE Instruct

GENERAL PURPOSE
Microsoft2024Mixture of Experts Transformer
Parameters
42B
Context
128K
Speed
15-30 tok/s
Pricing
$0.001/1K
Text
MMLU: 78.9% • HumanEval: 75.8%

Phi-4 14B

REASONING SPECIALIZED
Microsoft2024Advanced Transformer Decoder
Parameters
14B
Context
16K
Speed
30-60 tok/s
Pricing
$0.0015/1K
Text
MMLU: 84.7% • HumanEval: 82.6%

Pythia 12B

RESEARCH
EleutherAI2023Training-focused Transformer
Parameters
12B
Context
2K
Speed
40-80 tok/s
Pricing
Free
Text
MMLU: 47.2% • HumanEval: 13.2%

Qwen 72B Chat

GENERAL PURPOSE
Alibaba2023Instruction-tuned Transformer
Parameters
72B
Context
32K
Speed
12-25 tok/s
Pricing
Free
Text
MMLU: 77.4% • HumanEval: 64.6%

Qwen2.5 72B

GENERAL PURPOSE
Alibaba2024Enhanced Transformer Decoder
Parameters
72B
Context
128K
Speed
15-30 tok/s
Pricing
$0.0005/1K
Text
MMLU: 84.2% • HumanEval: 80.7%

Stable Code 3B

CODE SPECIALIZED
Stability AI2024Code-specialized Transformer
Parameters
3B
Context
16K
Speed
120-250 tok/s
Pricing
Free
Text
MMLU: 46.8% • HumanEval: 54.1%

StableLM Tuned Alpha 7B

GENERAL PURPOSE
Stability AI2023Instruction-tuned Transformer
Parameters
7B
Context
4K
Speed
80-150 tok/s
Pricing
Free
Text
MMLU: 42.9% • HumanEval: 20.9%

StarCoder 15.5B

CODE SPECIALIZED
Hugging Face2023Code-specialized Transformer
Parameters
15.5B
Context
8K
Speed
35-70 tok/s
Pricing
Free
Text
MMLU: 33.6% • HumanEval: 33.6%

StarCoder2 15B

CODE SPECIALIZED
Hugging Face2024Enhanced Code Transformer
Parameters
15B
Context
16K
Speed
40-80 tok/s
Pricing
Free
Text
MMLU: 46.2% • HumanEval: 46.2%

T5 11B

GENERAL PURPOSE
Google2019Text-to-Text Transformer
Parameters
11B
Context
512
Speed
20-40 tok/s
Pricing
Free
Text
MMLU: 68.7% • HumanEval: 26.2%

T5 Base

GENERAL PURPOSE
Google2019Text-to-Text Transformer
Parameters
220M
Context
512
Speed
150-250 tok/s
Pricing
Free
Text
MMLU: 52.4% • HumanEval: 15.8%

Titan Text Premier

GENERAL PURPOSE
Amazon2023Transformer Decoder
Parameters
~175B
Context
32K
Speed
15-35 tok/s
Pricing
$0.0005/1K
Text
MMLU: 75.2% • HumanEval: 42.8%

Turing-NLG 17B

RESEARCH
Microsoft2020Transformer Language Model
Parameters
17B
Context
2K
Speed
15-30 tok/s
Pricing
N/A
Text
MMLU: 45.8% • HumanEval: 12.4%

Vicuna 33B v1.3

CONVERSATIONAL
UC Berkeley, CMU, Stanford, UC San Diego, MBZUAI2023LLaMA-based Instruction-tuned
Parameters
33B
Context
2K
Speed
20-40 tok/s
Pricing
Free
Text
MMLU: 59.2% • HumanEval: 25.6%

o1

REASONING
OpenAI2024Reasoning-optimized Transformer
Parameters
Unknown
Context
200K
Speed
10-25 tok/s
Pricing
$0.015/1K
TextVision
MMLU: 94.8% • HumanEval: 92.3%

o1-mini

REASONING
OpenAI2024Reasoning-optimized Transformer
Parameters
Unknown
Context
128K
Speed
10-30 tok/s
Pricing
$0.003/1K
Text
MMLU: 85.2% • HumanEval: 87%

o1-preview

REASONING
OpenAI2024Reasoning-optimized Transformer
Parameters
Unknown
Context
128K
Speed
5-15 tok/s
Pricing
$0.015/1K
Text
MMLU: 90.8% • HumanEval: 89.7%