Cerebras pricing

Cerebras Inference serves a small set of open models on its wafer-scale chips, billed per token, with dedicated capacity by agreement.

custom-siliconserverlessdedicated

Verified 9 Nov 2026

gpt-oss-120b per 1M tokens: $0.35 in $0.75 out
Output speed vendor-stated: 3,000 tok/s

Go to Cerebras Compare with Cloudflare Workers AI, Replicate, SambaNova Request a quote

Every field, sourced

Marker opens the source

Status Whether the provider's public inference API is live.: Active 1
Hardware Chips that serve requests: GPUs, the vendor's own custom silicon, or both.: Custom silicon 1
Hardware detail Published chip types (e.g. LPU, wafer-scale engine, RDU, NVIDIA H100/B200).: Wafer-scale engine (CS-3 systems) 1
OpenAI API Accepts OpenAI Chat Completions-style requests, per vendor docs.: Yes 1
Serverless Pay-per-use API billed per token or per second, with no GPU to manage.: Yes 1
Dedicated Single-tenant deployments or reserved capacity.: Yes 1
Fine-tuning Hosted fine-tuning or LoRA adapters offered.: Not published
Models (published) models Model count as published or counted from the official model list on the verified date.: 2 2
Flagship models Notable models served, from the official list.: gpt-oss-120b, Qwen 3.8 27B 2
Serves gpt-oss-120b Offers the reference model gpt-oss-120b pay-per-token.: Yes 1
gpt-oss-120b in $/1M tok Standard serverless input price for the reference model gpt-oss-120b, per 1M tokens.: $0.35 1
gpt-oss-120b out $/1M tok Standard serverless output price for the reference model gpt-oss-120b, per 1M tokens.: $0.75 1
Llama 3.3 70B in $/1M tok Secondary reference: Llama 3.3 70B Instruct input price per 1M tokens, where still listed.: Not published
Llama 3.3 70B out $/1M tok Secondary reference: Llama 3.3 70B Instruct output price per 1M tokens, where still listed.: Not published
Vendor speed tok/s Vendor-stated output speed on gpt-oss-120b, linked to the vendor’s own page.: 3,000 1
Speed source Who published the speed figure and when.: Cerebras pricing and models overview ('~3000 tokens/s'), retrieved 2026-11-09 (vendor figure) 1
Dedicated H100 $/GPU-hr On-demand price of one H100 for dedicated deployments.: Not published
Regions Published serving or data regions.: US data centers 1
SOC 2 SOC 2 status as stated by the vendor.: Not published
HIPAA Vendor states HIPAA support or offers a BAA.: Not published
ZDR Zero data retention: the vendor states prompts and outputs are not stored, by default or as a setting.: Yes 3
Free tier Free credits or free rate-limited access, as published.: $5 one-time promo credit plus free dev tier 1

Pricing detail

gpt-oss-120b $0.35 in / $0.75 out per 1M (up from $0.25/$0.69 at Aug 2025 launch), ~3,000 tok/s (vendor). Llama 3.3 70B no longer listed. Privacy policy: inference inputs/outputs not retained.

Sources

Cerebras Pricing accessed 2026-11-09
Models overview accessed 2026-11-09
Privacy Policy accessed 2026-11-09

Get dedicated or volume pricing from Cerebras and similar providers

Tell us what you need once. We send it to matching providers and they reply to you directly.