The best open LLMs for your use case:

1GLM-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:

#1

FrontierSWE

Coding Agents

74.4
#1

Terminal-Bench 2.1

Coding Agents

81
#1

DeepSWE

Coding Agents

46.2
#1

SWE-Bench Pro

Coding Agents

62.1
#1

HLE

General Knowledge

40.5
#1

MCP-Atlas

Agents and Function Calling

76.8
#1

EQBench

Creative Writing

1575
#2

GPQA-Diamond

General Knowledge

91.2
#1

FrontierSWE

Coding Agents

74.4
#1

Terminal-Bench 2.1

Coding Agents

81
#1

DeepSWE

Coding Agents

46.2
#1

SWE-Bench Pro

Coding Agents

62.1
#1

HLE

General Knowledge

40.5
#1

MCP-Atlas

Agents and Function Calling

76.8
#1

EQBench

Creative Writing

1575
#2

GPQA-Diamond

General Knowledge

91.2
Try it out
2Kimi K2.7 CodeMoonshot

Coding-specialized variant of Moonshot's Kimi K2.7 with long-context support and strong agentic software engineering performance.

Speed:

Intelligence:

Price: (1M Tokens)

$0.95 / 4.00

Cached input: (1M Tokens)

$0.19

Context: (tokens)

262,144

Inputs:

ImageText

Benchmarks:

#1

MLS Bench Lite

Coding Agents

35.1
#1

Program Bench

Coding Agents

53.6
#1

Kimi Code Bench v2

Coding Agents

62
#1

MCP-Mark Verified

Agents and Function Calling

81.1
#1

Kimi Claw 24/7 Bench

Agents and Function Calling

46.9
#2

MCP-Atlas

Agents and Function Calling

76
#1

MLS Bench Lite

Coding Agents

35.1
#1

Program Bench

Coding Agents

53.6
#1

Kimi Code Bench v2

Coding Agents

62
#1

MCP-Mark Verified

Agents and Function Calling

81.1
#1

Kimi Claw 24/7 Bench

Agents and Function Calling

46.9
#2

MCP-Atlas

Agents and Function Calling

76
Try it out

Use case:

Coding Agents

Features:

Low Latency