Llama-3.3-70B

Llama-3.3-70B is the latest version of the popular Llama 3 series, optimized for dialogue and assistant-like scenarios. The model has been trained on over 15 trillion tokens from open-source data, with an expanded context window of up to 128 thousand tokens, enabling it to efficiently handle long conversations and large texts. Llama-3.3-70B utilizes supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to better adapt to user queries, providing not only accuracy but also safer responses.

The model is distributed under the special commercial Llama 3.3 Community License, which allows its use in both research and commercial applications, provided the acceptable usage policy is followed. However, Llama-3.3-70B is not intended for use in mission-critical domains (e.g., healthcare, finance, infrastructure control, etc.).

Llama-3.3-70B is a powerful language model optimized for conversational and assistant tasks. It's ideal for building chatbots and automating customer support. In essence, it serves as a versatile tool for natural language generation, summarization, translation, code writing, and a wide range of other applications requiring high-quality language processing.


Announce Date: 06.12.2024
Parameters: 70B
Context: 131K
Attention Type: Full Attention
VRAM requirements: 72.6 GB using 4 bits quantization
Developer: Meta AI
Transformers Version: 4.48.0
License: LLAMA 3.3

Public endpoint

Use our pre-built public endpoints to test inference and explore Llama-3.3-70B capabilities.
Model Name Context Type GPU TPS Status Link
There are no public endpoints for this model yet.

Private server

Rent your own physically dedicated instance with hourly or long-term monthly billing.

We recommend deploying private instances in the following scenarios:

  • maximize endpoint performance,
  • enable full context for long sequences,
  • ensure top-tier security for data processing in an isolated, dedicated environment,
  • use custom weights, such as fine-tuned models or LoRA adapters.

Recommended configurations for hosting Llama-3.3-70B

Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
teslaa10-4.16.128.160 16 131072 160 4 $1.75 Launch
teslaa100-1.16.128.160 16 131072 160 1 $2.71 Launch
rtx3090-4.16.128.160 16 131072 160 4 $3.23 Launch
rtx4090-4.16.128.160 16 131072 160 4 $4.26 Launch
rtx5090-3.16.96.160 16 98304 160 3 $4.34 Launch
teslah100-1.16.128.160 16 131072 160 1 $5.23 Launch
Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
teslaa100-2.24.256.160 24 262144 160 2 $5.35 Launch
rtx5090-4.16.128.160 16 131072 160 4 $5.74 Launch
teslah100-2.24.256.160 24 262144 160 2 $10.40 Launch
Prices:
Name vCPU RAM, MB Disk, GB GPU Price, hour
teslaa100-3.32.384.240 32 393216 240 3 $8.00 Launch
rtx4090-8.44.256.240 44 262144 240 8 $8.59 Launch
rtx5090-6.44.256.240 44 262144 240 6 $8.86 Launch
teslah100-3.32.384.240 32 393216 240 3 $15.58 Launch

Related models

Need help?

Contact our dedicated neural networks support team at nn@immers.cloud or send your request to the sales department at sale@immers.cloud.