TAU2
Measured May 14, 2026Source
Score
0.27
Models15 2 
Meta
Llama 3.3 Instruct 70B
MetaMetaUnited States
Llama 3.3 Instruct 70B is a large, open-source instruction-tuned model from Meta, optimized for complex reasoning, code generation, and following detailed instructions. It features a 128K token context window and delivers strong performance on various benchmarks, positioning it as a powerful and versatile general-purpose assistant.
ReasoningCodingLong contextMultimodal
Input / 1M tokens
$0.585Output / 1M tokens
$0.71Output tokens/s
90.82First-token seconds
0.6sSupported plans
6Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.03
Lcr
Measured May 14, 2026Source
Score
0.15
Ifbench
Measured May 14, 2026Source
Score
0.47
Aime 25
Measured May 14, 2026Source
Score
0.08
Aime
Measured May 14, 2026Source
Score
0.3
Math 500
Measured May 14, 2026Source
Score
0.77
Scicode
Measured May 14, 2026Source
Score
0.26
Livecodebench
Measured May 14, 2026Source
Score
0.29
Hle
Measured May 14, 2026Source
Score
0.04
Gpqa
Measured May 14, 2026Source
Score
0.5
Mmlu Pro
Measured May 14, 2026Source
Score
0.71
Artificial Analysis Math Index
Measured May 14, 2026Source
Score
7.7
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
10.7
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
14.5
Plan availability
Products and plans that support this model
io.net Plan
io.net Plan
io.net is a decentralized GPU cloud platform providing subscription plans and pay-as-you-go access to a variety of AI models for chat, coding, content creation, and automation workflows.
Synthetic
Synthetic
Synthetic is a platform that runs open-source AI models in private, secure datacenters. It offers subscription and usage-based pricing for accessing models like DeepSeek, Llama, MiniMax, Kimi, and others, with OpenAI-compatible API access for use in coding agents and tools.
Discussion
Thinking... Make sure you are connected to GitHub server