macrostack

AI API price index · the same model at every host · read 10 Nov 2026 to 17 Nov 2026

gpt-oss-120b API pricing at every host

10 hosts sell gpt-oss-120b at a published per-token price. Blended, it costs $0.10 to $0.45 per million tokens — 4.5× apart for the same model. Cheapest: Novita AI. Each price read on the host’s own pricing page and linked to it.

How this is ranked

The same model, so a cheapest is honest. Hosts are ordered by blended price, (3 × input + 1 × output) ÷ 4 per million tokens, then by output, then input, then name. Equal prices are shown as equal, never as one host winning.

The index decides what counts. Standard tier, global rate, list price, standard prompt length — the same rules as the AI API price index. Every other price for gpt-oss-120b stays on this page and says why it is not ranked. A page needs 3 ranked hosts to exist at all.

gpt-oss-120b at 10 hosts, cheapest blended first

Per 1M tokens, read 10 Nov 2026 to 17 Nov 2026.

HostInputCached inputOutputBlendedWhat the host’s page records
Novita AIOpenAI GPT OSS 120B$0.05—$0.25$0.10Tier: serverless · Context 128Kread 17 Nov 2026
Google Vertex AI (Agent Platform)gpt-oss-120b$0.09—$0.36$0.1575—read 17 Nov 2026
Amazon Bedrockgpt-oss-120b$0.15—$0.60$0.2625On-demand, in-Region (us-east-1)read 14 Nov 2026
Fireworks AIOpenAI GPT OSS 120B$0.15$0.015$0.60$0.2625Tier: serverless · Standard mode; Priority $0.18 / $0.018 / $0.72read 10 Nov 2026
GroqGPT OSS 120B (openai/gpt-oss-120b)$0.15$0.075$0.60$0.2625—read 10 Nov 2026
Nebius Token Factorygpt-oss-120b$0.15$0.10$0.60$0.2625public endpoint (shared; the page says suitable for testing, not production)read 17 Nov 2026
Together AIgpt-oss-120B$0.15—$0.60$0.2625Tier: serverlessread 10 Nov 2026
SambaNova Cloudgpt-oss-120b$0.22—$0.59$0.3125—read 17 Nov 2026
CerebrasGPT OSS 120B$0.35—$0.75$0.45Tier: developer · Developer tier: "For development, evaluation, and experimentation; not intended for production use." About 3,000 tokens a second; context 65k on the free trial, 131k paidread 17 Nov 2026
Cloudflare Workers AI@cf/openai/gpt-oss-120b$0.35—$0.75$0.45—read 17 Nov 2026

Blended = (3 × input + 1 × output) ÷ 4. A dash means the host prints no cached-input price for this model (Novita AI, Google Vertex AI (Agent Platform), Amazon Bedrock, Together AI, SambaNova Cloud, Cerebras and Cloudflare Workers AI).

Where the hosts differ

Only what each host's own price row records. Nothing here is inferred.

  • Context length: Novita AI, Context 128K; Cerebras, context 65k on the free trial, 131k paid. The other 8 hosts' rows do not record it.
  • Speed: Cerebras, About 3,000 tokens a second.
  • Quantisation: no host's price row records it — ask before assuming two endpoints run identical weights.
  • Use: Nebius Token Factory, "suitable for testing, not production"; Cerebras, "not intended for production use".

Read next

Questions

How much does the gpt-oss-120b API cost?
Read on each host's own page (10 Nov 2026 to 17 Nov 2026), gpt-oss-120b costs from $0.10 to $0.45 per million tokens blended at 10 hosts — 4.5× apart for the same model. The cheapest is Novita AI at $0.05 in and $0.25 out per million tokens; the dearest is Cloudflare Workers AI at $0.35 in and $0.75 out per million tokens.
Which host is cheapest for gpt-oss-120b?
Novita AI, at $0.10 per million tokens blended ($0.05 in and $0.25 out per million tokens). The other 9 hosts charge more for the same model.
Is gpt-oss-120b the same model at every host?
The model is the same open-weight release, matched by name on each host's price list. What the hosts' own price rows record differs: context length (Novita AI: Context 128K; Cerebras: context 65k on the free trial, 131k paid), speed (Cerebras: About 3,000 tokens a second), use (Nebius Token Factory: "suitable for testing, not production"; Cerebras: "not intended for production use"). None of these price rows records quantisation, so ask the host before assuming two endpoints run identical weights.
What does blended price mean here?
Blended = (3 × input + 1 × output) ÷ 4, per million tokens: three input tokens to every output token, a common mix for chat and retrieval. It ranks hosts on one number; the input, cached-input and output prices are all on this page for any other mix.
Can I use gpt-oss-120b free?
Cloudflare Workers AI (free tier) and Groq (free tier) name it among their free use; the limits are on our free AI API page. Cerebras and Fireworks AI also give a one-time starting credit that could be spent on it.

Prices read 10 Nov 2026 to 17 Nov 2026 on each host’s own pricing page. Macrostack measures nothing and is paid by no host on this page. Part of the AI API price index on Layer 2.5 — Compute.

The Macrostack brief

New swaps, worth your inbox.

A short, occasional email when we add a high-intent alternative or ship a new head-to-head. No spam, no selling your address — unsubscribe in one click.