COMPARE.

Unspecified

llamagate/openthinker-7b

Models^*

Maximum output

8,192 tokens^*

Maximum input

32,768 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

llamagate/qwen3-vl-8b

Models^*

Maximum output

8,192 tokens^*

Maximum input

32,768 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

meta.llama3-1-405b-instruct-v1:0

Models^*

Maximum output

4,096 tokens^*

Maximum input

128,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

meta.llama3-2-1b-instruct-v1:0

Models^*

Maximum output

4,096 tokens^*

Maximum input

128,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

meta.llama3-2-3b-instruct-v1:0

Models^*

Maximum output

4,096 tokens^*

Maximum input

128,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

mistral.mistral-small-2402-v1:0

Models^*

Maximum output

4,096 tokens^*

Maximum input

32,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

mistral/codestral-mamba-latest

Models^*

Maximum output

256,000 tokens^*

Maximum input

256,000 tokens^*

API mode

Chat^*

Tool choice

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

mistral/labs-leanstral-1-5-1

Models^*

Maximum output

131,072 tokens^*

Maximum input

262,144 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

moonshot/moonshot-v1-128k

Models^*

Maximum output

131,072 tokens^*

Maximum input

131,072 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

moonshot/moonshot-v1-32k

Models^*

Maximum output

32,768 tokens^*

Maximum input

32,768 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

moonshot/moonshot-v1-8k

Models^*

Maximum output

8,192 tokens^*

Maximum input

8,192 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

moonshot/moonshot-v1-auto

Models^*

Maximum output

131,072 tokens^*

Maximum input

131,072 tokens^*

Tool calling

Yes^*

Structured output

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/deepseek-ai/DeepSeek-V4.1-Flash

Models^*

Maximum output

384,000 tokens^*

Maximum input

1,048,000 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/MiniMaxAI/MiniMax-M3

Models^*

Maximum output

1,048,576 tokens^*

Maximum input

1,048,576 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/moonshotai/Kimi-K2.6

Models^*

Maximum output

262,144 tokens^*

Maximum input

262,144 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/moonshotai/Kimi-K2.7-Code

Models^*

Maximum output

262,144 tokens^*

Maximum input

262,144 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/moonshotai/Kimi-K3

Models^*

Maximum output

1,024,000 tokens^*

Maximum input

1,024,000 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/nvidia/Cosmos3-Super-Reasoner

Models^*

Maximum output

262,144 tokens^*

Maximum input

262,144 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/nvidia/Nemotron-3-Nano-Omni

Models^*

Maximum output

262,144 tokens^*

Maximum input

262,144 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

nebius/zai-org/GLM-5.3-Flash

Models^*

Maximum output

1,024,000 tokens^*

Maximum input

1,024,000 tokens^*

Reasoning

Yes^*

Tool calling

Yes^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

novita/baidu/ernie-4.5-300b-a47b-paddle

Models^*

Maximum output

12,000 tokens^*

Maximum input

123,000 tokens^*

Structured output

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

novita/baidu/ernie-4.5-vl-424b-a47b

Models^*

Maximum output

16,000 tokens^*

Maximum input

123,000 tokens^*

Reasoning

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

novita/deepseek/deepseek_v3

Models^*

Maximum output

16,000 tokens^*

Maximum input

64,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence

Unspecified

novita/deepseek/deepseek-v3/community

Models^*

Maximum output

8,000 tokens^*

Maximum input

64,000 tokens^*

Tool calling

Yes^*

API mode

Chat^*

Radar · unrated0/4 core tests · not enough evidence