AI21 Labs vs Lepton AI: Token Pricing, Speed & Intelligence
Full comparison of AI21 Labs and Lepton AI — live token pricing, latency, throughput, context window, strengths, weaknesses, and best use cases. Rates sourced from public APIs — verify before use.
AI21 Labs
Jamba 1.6 — hybrid SSM-Transformer models with 256K context
AI21 Labs develops the Jamba model family, which uses a hybrid SSM (Mamba) and Transformer architecture. Jamba 1.6 Large and Mini offer 256K token context windows at competitive prices. The hybrid architecture delivers faster inference and lower memory usage than pure-Transformer models of comparable size.
Lepton AI
NVIDIA DGX Cloud Lepton — enterprise serverless inference on DGX infrastructure
Lepton AI was acquired by NVIDIA and rebranded as NVIDIA DGX Cloud Lepton in 2026. The platform now runs on NVIDIA's DGX Cloud infrastructure, offering serverless inference for popular open-weight models with enterprise-grade reliability. The OpenAI-compatible API and developer experience remain intact, now backed by NVIDIA's global DGX Cloud footprint.
Key metric comparison
—
—
—
—
—
—
—
—
—
—
—
—
Live token pricing
Strengths and weaknesses
AI21 Labs
Lepton AI
Key differentiators
Jamba's hybrid SSM-Transformer architecture enables 256K context at lower memory and latency cost than pure-Transformer models — with Jamba 1.6 Mini among the cheapest 256K-context options available.
The only serverless inference platform running directly on NVIDIA DGX Cloud — enterprise reliability with the latest NVIDIA hardware, no infrastructure management required.
Frequently asked questions
AI21 Labs FAQs
What is the Jamba architecture?
Jamba uses a hybrid of Mamba (SSM) and Transformer layers. This reduces memory usage and improves throughput on long sequences compared to pure-Transformer models, while maintaining strong reasoning performance.
How much does AI21 Labs API cost?
Jamba 1.6 Large is $2.00/$8.00 per 1M input/output tokens. Jamba 1.6 Mini is $0.20/$0.40 per 1M tokens. Both support 256K context. See ai21.com/pricing for current rates.
What context window does Jamba 1.6 support?
Both Jamba 1.6 Large and Jamba 1.6 Mini support 256K token context windows, making them suitable for long-document analysis and extended conversations.
Lepton AI FAQs
What happened to Lepton AI?
Lepton AI was acquired by NVIDIA and rebranded as NVIDIA DGX Cloud Lepton in 2026. The platform now runs on NVIDIA's DGX Cloud infrastructure with the same developer-friendly API.
What models does NVIDIA DGX Cloud Lepton support?
DGX Cloud Lepton hosts Llama 3.3 70B and other popular open-weight models. The catalog is focused on widely-used production models.
Is DGX Cloud Lepton good for production workloads?
Yes. Backed by NVIDIA's DGX Cloud infrastructure, DGX Cloud Lepton offers enterprise-grade reliability and SLAs suitable for production workloads.
Provider resources
AI21 Labs — Jamba 1.6 — hybrid SSM-Transformer models with 256K context
AI21 Labs develops the Jamba model family, which uses a hybrid SSM (Mamba) and Transformer architecture. Jamba 1.6 Large and Mini offer 256K token context windows at competitive prices. The hybrid architecture delivers faster inference and lower memory usage than pure-Transformer models of comparable size.
Jamba's hybrid SSM-Transformer architecture enables 256K context at lower memory and latency cost than pure-Transformer models — with Jamba 1.6 Mini among the cheapest 256K-context options available.
Lepton AI — NVIDIA DGX Cloud Lepton — enterprise serverless inference on DGX infrastructure
Lepton AI was acquired by NVIDIA and rebranded as NVIDIA DGX Cloud Lepton in 2026. The platform now runs on NVIDIA's DGX Cloud infrastructure, offering serverless inference for popular open-weight models with enterprise-grade reliability. The OpenAI-compatible API and developer experience remain intact, now backed by NVIDIA's global DGX Cloud footprint.
The only serverless inference platform running directly on NVIDIA DGX Cloud — enterprise reliability with the latest NVIDIA hardware, no infrastructure management required.
Key strengths compared
AI21 Labs
- Hybrid SSM-Transformer architecture delivers faster inference than pure-Transformer models
- 256K token context window on both Jamba 1.6 Large and Mini
- Jamba 1.6 Mini at $0.20/$0.40 per 1M tokens is highly cost-competitive
- Smaller model family than frontier labs — two models vs dozens at OpenAI or Anthropic
- Less ecosystem support than OpenAI-compatible providers
Lepton AI
- Backed by NVIDIA DGX Cloud infrastructure
- Enterprise-grade reliability and SLAs
- Serverless — no infrastructure management
- Smaller model catalog than Together AI or Fireworks
- Pricing may be higher than budget inference providers
Key weaknesses compared
AI21 Labs
- Smaller model family than frontier labs — two models vs dozens at OpenAI or Anthropic
- Less ecosystem support than OpenAI-compatible providers
- No image, audio, or multimodal models
Lepton AI
- Smaller model catalog than Together AI or Fireworks
- Pricing may be higher than budget inference providers
- No fine-tuning support
Provider category context
AI21 Labs is a frontier lab, founded in 2017. Lepton AI is a inference api, founded in 2023. AI21 Labs as a frontier lab trains and serves its own proprietary models. Lepton AI as an inference API provider hosts open-weight models — typically offering lower prices for equivalent capability tiers but without access to proprietary frontier models.
How to choose between them
Choose AI21 Labs if you need Long-context tasks. Choose Lepton AI if you need backed by nvidia dgx cloud infrastructure. For high-volume production workloads, run a cost comparison using the token pricing table above with your actual prompt/completion token ratio — the cheapest provider depends heavily on your input-to-output token ratio.