The best open LLMs for your use case:

1Kimi K3Moonshot

Moonshot AI's 2.8-trillion-parameter open-weight frontier model with native vision, 1M-token context, and state-of-the-art agentic coding and tool use.

Speed:

Intelligence:

Price: (1M Tokens)

$3.00 / 15.00

Cached input: (1M Tokens)

$0.30

Context: (tokens)

1,048,576

Inputs:

ImageText

Benchmarks:

#1

Program Bench

Coding Agents

77.8
#1

DeepSWE

Coding Agents

69
#1

FrontierSWE

Coding Agents

81.2
#1

Terminal-Bench 2.1

Coding Agents

88.3
#1

GPQA-Diamond

General Knowledge

93.5
#1

HLE

General Knowledge

43.5
#1

MMMU-Pro

Multimodal - Vision

81.6
#1

MCP-Atlas

Agents and Function Calling

84.2
#1

Program Bench

Coding Agents

77.8
#1

DeepSWE

Coding Agents

69
#1

FrontierSWE

Coding Agents

81.2
#1

Terminal-Bench 2.1

Coding Agents

88.3
#1

GPQA-Diamond

General Knowledge

93.5
#1

HLE

General Knowledge

43.5
#1

MMMU-Pro

Multimodal - Vision

81.6
#1

MCP-Atlas

Agents and Function Calling

84.2
Try it out
2GLM-5.2Z.ai

Agentic-focused refinement of GLM-5.1 from Z.ai with improved coding, tool use, and reasoning, plus extended 256K context.

Speed:

Intelligence:

Price: (1M Tokens)

$1.40 / 4.40

Cached input: (1M Tokens)

$0.26

Context: (tokens)

262,144

Inputs:

ImageText

Benchmarks:

#2

FrontierSWE

Coding Agents

74.4
#2

Terminal-Bench 2.1

Coding Agents

81
#2

DeepSWE

Coding Agents

46.2
#1

SWE-Bench Pro

Coding Agents

62.1
#1

EQBench

Creative Writing

1575
#2

HLE

General Knowledge

40.5
#2

MCP-Atlas

Agents and Function Calling

76.8
#3

GPQA-Diamond

General Knowledge

91.2
#2

FrontierSWE

Coding Agents

74.4
#2

Terminal-Bench 2.1

Coding Agents

81
#2

DeepSWE

Coding Agents

46.2
#1

SWE-Bench Pro

Coding Agents

62.1
#1

EQBench

Creative Writing

1575
#2

HLE

General Knowledge

40.5
#2

MCP-Atlas

Agents and Function Calling

76.8
#3

GPQA-Diamond

General Knowledge

91.2
Try it out

Use case:

Coding Agents