Llama-3.3-70B-Instruct

开源权重

Meta · Llama · meta/llama-3.3-70b-instruct

价格 $0.05 起 ~ $2.92 / 百万 tokens · 共 21 家算力服务商部署

各服务商对比

服务商模型 ID上下文最大输出输入 $/M输出 $/M缓存 $/MAPI 端点

NanoGPT meta-llama/llama-3.3-70b-instruct 131.1K 16.4K $0.05 $0.23 — nano-gpt.com ↗

Cortecs llama-3.3-70b-instruct 131K 131K $0.09 $0.28 — api.cortecs.ai ↗

Kilo Gateway meta-llama/llama-3.3-70b-instruct 131.1K 16.4K $0.1 $0.32 — api.kilo.ai ↗

Meganova meta-llama/Llama-3.3-70B-Instruct 131.1K 16.4K $0.1 $0.3 — api.meganova.ai ↗

OpenRouter meta-llama/llama-3.3-70b-instruct 131.1K 16.4K $0.1 $0.32 — openrouter.ai ↗

Nebius Token Factory meta-llama/Llama-3.3-70B-Instruct 128K 8.2K $0.13 $0.4 $0.01 api.tokenfactory.nebius.com ↗

Helicone llama-3.3-70b-instruct 128K 16.4K $0.13 $0.39 — ai-gateway.helicone.ai ↗

LLM Gateway llama-3.3-70b-instruct 131.1K 4.1K $0.13 $0.4 — api.llmgateway.io ↗

IO.NET meta-llama/Llama-3.3-70B-Instruct 128K 4.1K $0.13 $0.38 $0.07 api.intelligence.io.solutions ↗

NovitaAI meta-llama/llama-3.3-70b-instruct 131.1K 120K $0.14 $0.4 — api.novita.ai ↗

Hugging Face meta-llama/Llama-3.3-70B-Instruct 131.1K 4.1K $0.59 $0.79 — router.huggingface.co ↗

Regolo AI llama-3.3-70b-instruct 128K 16.4K $0.6 $2.7 — api.regolo.ai ↗

Azure llama-3.3-70b-instruct 128K 32.8K $0.71 $0.71 — —

Weights & Biases meta-llama/Llama-3.3-70B-Instruct 128K 128K $0.71 $0.71 — api.inference.wandb.ai ↗

Azure Cognitive Services llama-3.3-70b-instruct 128K 32.8K $0.71 $0.71 — —

Scaleway llama-3.3-70b-instruct 100K 16.4K $0.9 $0.9 — api.scaleway.ai ↗

Berget.AI meta-llama/Llama-3.3-70B-Instruct 128K 8.2K $0.99 $0.99 — api.berget.ai ↗

CloudFerro Sherlock meta-llama/Llama-3.3-70B-Instruct 70K 70K $2.92 $2.92 — api-sherlock.cloudferro.com ↗

GitHub Models meta/llama-3.3-70b-instruct 128K 32.8K 免费免费 — models.github.ai ↗

Nvidia meta/llama-3.3-70b-instruct 128K 4.1K 免费免费 — integrate.api.nvidia.com ↗

Llama llama-3.3-70b-instruct 128K 4.1K 免费免费 — api.llama.com ↗

上下文窗口128K

最大输出4.1K

知识截止2023-12

发布日期2024-12-06

系列llama

最后更新2024-12-06

输入模态文本

输出模态文本

能力工具调用温度调节文件附件

权重下载 Hugging Face ↗

基准跑分 Artificial Analysis Coding Index 10.7 (index) · SciCode 26 (percent correct) · Terminal-Bench Hard 3 (success rate)

Muse Spark 1.1 Llama 4 Scout 17B Instruct Llama 4 Maverick 17B Instruct

数据来自开源项目 models.dev · 价格为美元 / 百万 tokens,以各服务商官方公示为准。