The best open LLMs for your use case:

1Gemma 4 31BGoogle

Google's open multimodal model with strong multilingual coverage, 256K context, and text + image input.

Speed:

Intelligence:

Price: (1M Tokens)

$0.39 / 0.97

Context: (tokens)

262,144

Inputs:

ImageText

Benchmarks:

#2

Multilingual MMLU

Multilingual

88.4
#1

MRCR 128k

Summarization

66.4
#3

MMLU-Pro

General Knowledge

85.2
#3

TAU2-Bench

Agents and Function Calling

76.9
#4

LiveCodeBench

Coding Agents

80
#4

MMMU-Pro

Multimodal - Vision

76.9
#5

HLE

General Knowledge

19.5
#6

GPQA-Diamond

General Knowledge

84.3
#2

Multilingual MMLU

Multilingual

88.4
#1

MRCR 128k

Summarization

66.4
#3

MMLU-Pro

General Knowledge

85.2
#3

TAU2-Bench

Agents and Function Calling

76.9
#4

LiveCodeBench

Coding Agents

80
#4

MMMU-Pro

Multimodal - Vision

76.9
#5

HLE

General Knowledge

19.5
#6

GPQA-Diamond

General Knowledge

84.3
Try it out
2Kimi K2.6Moonshot

1T-parameter MoE flagship from Moonshot with long-horizon coding, agent swarms scaling to 300 sub-agents, and state-of-the-art reasoning.

Speed:

Intelligence:

Price: (1M Tokens)

$1.20 / 4.50

Cached input: (1M Tokens)

$0.20

Context: (tokens)

262,144

Inputs:

ImageText

Benchmarks:

#1

SciCode

Coding Agents

52.2
#1

MCP-Mark

Agents and Function Calling

55.9
#1

MMMU-Pro

Multimodal - Vision

79.4
#1

Apex Agents

Agents and Function Calling

27.9
#1

FrontierCode

Coding Agents

3.8
#2

LiveCodeBench

Coding Agents

89.6
#2

Terminal-Bench 2.0

Coding Agents

66.7
#2

Claw-Eval

Agents and Function Calling

62.3
#1

SciCode

Coding Agents

52.2
#1

MCP-Mark

Agents and Function Calling

55.9
#1

MMMU-Pro

Multimodal - Vision

79.4
#1

Apex Agents

Agents and Function Calling

27.9
#1

FrontierCode

Coding Agents

3.8
#2

LiveCodeBench

Coding Agents

89.6
#2

Terminal-Bench 2.0

Coding Agents

66.7
#2

Claw-Eval

Agents and Function Calling

62.3
Try it out

Use case:

Multilingual

Features:

Low Latency