MetaactiveOpen Source

Llama 3.3 70B

llama-3.3-70b

High-performance open-source model competitive with larger proprietary models.

Context Window

131.1K

tokens

Max Output

8.2K

tokens

Input Price

—

per 1M tokens

Output Price

—

per 1M tokens

Details

Familyllama-3

Parameters70B

Training Cutoff2024-12-01

ReleasedDecember 6, 2024

Aliasesmeta-llama/Llama-3.3-70B-Instruct

Capabilities

FunctionsStreamingJSON ModeCodeTool Use

Documentation

Evaluation Scores(4 benchmarks)

HumanEvalFunction-level Python code generation

74.8%

MMLU-ProHarder successor to MMLU

62.5%

MATH-500Competition-style math

58%

GPQA DiamondPhD-level science questions

42%

Quick Access

curl pikaainews.com/api/models/meta-llama-3-3-70b

npx pika-models info meta-llama-3-3-70b

Get API Access

Official

AWS Bedrock

Amazon Bedrock. Official partnerships with Anthropic, Meta, Mistral, Cohere.

Third-Party Providers & Aggregators

Cerebras

Wafer-scale inference. 1000+ tokens/sec for select models.

DeepInfra

Lowest per-token rates for open-source models.

Fireworks AI

Fastest inference engine. Multimodal support, HIPAA/SOC2.

Groq

Ultra-fast LPU inference. Best latency for real-time apps.

OpenRouter

500+ models, one API key. Pay-per-token, no minimums.

SiliconFlow

China-optimized inference. Strong Qwen/DeepSeek support.

Together AI

Fast open-source model inference. Sub-100ms latency.