The best LLMs for your use case:

1Kimi K2 Instruct 0905Moonshot

Advanced multimodal AI model from Moonshot with 256k context length for instruction following.

Speed:

Intelligence:

Price: (1M Tokens)

$1.20

Inputs:

ImageText

JSON Mode:

Function Calling:

Benchmarks:

#1

EQBench

Creative Writing

1565.3
#4

LMArena

Chat

1421
#5

WebDevArena

Code

1314
#5

BFCL

Agents and Function Calling

71.1
#6

LongBenchv2

Summarization

44.9
#7

GPQA-Diamond

General Knowledge

75.1
#7

SimpleQA

General Knowledge

31
#11

MMLU-Pro

General Knowledge

81.1
#14

LiveCodeBench

Code

53.7
#1

EQBench

Creative Writing

1565.3
#4

LMArena

Chat

1421
#5

WebDevArena

Code

1314
#5

BFCL

Agents and Function Calling

71.1
#6

LongBenchv2

Summarization

44.9
#7

GPQA-Diamond

General Knowledge

75.1
#7

SimpleQA

General Knowledge

31
#11

MMLU-Pro

General Knowledge

81.1
#14

LiveCodeBench

Code

53.7
Try it out
2Qwen3 235B-A22B Instruct 2507Qwen

Qwen's 235B parameter instruct model with 22B active parameters, optimized for high-throughput instruction following and general tasks.

Speed:

Intelligence:

Price: (1M Tokens)

$0.25

Inputs:

ImageText

JSON Mode:

Function Calling:

Benchmarks:

#2

EQBench

Creative Writing

1275
#2

SimpleQA

General Knowledge

54.3
#6

LMArena

Chat

1419
#6

Multilingual MMLU

Multilingual

79.4
#7

WebDevArena

Code

1189
#7

BFCL

Agents and Function Calling

70.9
#8

MMLU-Pro

General Knowledge

83
#15

LiveCodeBench

Code

51.8
#2

EQBench

Creative Writing

1275
#2

SimpleQA

General Knowledge

54.3
#6

LMArena

Chat

1419
#6

Multilingual MMLU

Multilingual

79.4
#7

WebDevArena

Code

1189
#7

BFCL

Agents and Function Calling

70.9
#8

MMLU-Pro

General Knowledge

83
#15

LiveCodeBench

Code

51.8
Try it out

Use case:

Creative Writing