Best LLM Inference Providers

Vendor profileChecked Nov 8, 2026

Cerebras

Cerebras Systems Inc. · cerebras.ai · Last checked:

  • Serverless per-token API
  • Dedicated endpoints (enterprise)
  • Custom inference hardware (wafer-scale)

Company status

Active, publicly listed (Nasdaq: CBRS); Class A common stock began trading May 14, 2026.

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Capabilities

  • Inference API on Cerebras wafer-scale hardware for open-weight models (Developer tier lists GPT OSS 120B and Qwen 3.8 27B)
  • OpenAI-compatible API at api.cerebras.ai/v1 for many common workflows
  • Dedicated endpoints for enterprise customers, with custom weights uploaded through a Management API
  • Fine-tuning and custom model weights by agreement on the Enterprise tier
  • Publishes approximate tokens per second per model on its pricing page (vendor claim)

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Who it tends to fit

  • Teams that want an API on non-GPU inference hardware for supported open-weight models
  • Enterprises that need dedicated endpoints with their own weights for a supported architecture

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Pricing

Published per-token pricing on https://www.cerebras.ai/pricing for the Developer tier (for example, GPT OSS 120B listed at $0.35 per 1M input tokens and $0.75 per 1M output tokens). Enterprise tier is quoted.

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Published certifications

Listed as stated by the vendor. Scope can differ by product or subsidiary; request current reports.

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Limitations to check

  • Developer tier is described as for development, evaluation, and experimentation, not production use
  • Small self-serve model list; dedicated endpoints and custom weights are enterprise-only
  • Custom weights must match the architecture the dedicated endpoint is provisioned for
  • Trust Center lists SOC 2 without stating report type publicly; request access for documents

Sources 1, 2, 3, 4, 5 · checked 2026-11-08

Sources

  1. Pricing page: Developer and Enterprise tiers; Developer tier per-token prices; developer tier not intended for production; custom weights and fine-tuning by agreementwww.cerebras.ai/pricingChecked 2026-11-08
  2. Trust Center compliance lists SOC 2, CCPA, GDPR; documents by requesttrust.cerebras.aiChecked 2026-11-08
  3. Cerebras API compatible with OpenAI client libraries for many common workflows; base URL https://api.cerebras.ai/v1inference-docs.cerebras.ai/resources/openaiChecked 2026-11-08
  4. Dedicated endpoints available to enterprise customers; custom weights must be compatible with the provisioned architectureinference-docs.cerebras.ai/dedicated/overviewChecked 2026-11-08
  5. IPO closing: Class A common stock began trading on Nasdaq Global Select Market on May 14, 2026 under CBRSwww.cerebras.ai/press-release/cerebras-systems-announces-closing-of-initial-public-offeringChecked 2026-11-08

Report a correction: [email protected]. Facts are re-checked at least every 90 days.