Arbitrage · gpt-oss-120b5 providers · $0.150 → $0.400 · 63% spread
Cheapest Provider for gpt-oss-120b
OpenAI's larger open release · reasoning-capable · still cheap on optimized stacks.
Cheapest input
Groq
$0.150/M
Speed leader
Fastest
Cerebras
850 tok/s
Wafer-scale chip
Savings calculator
Save 63%
vs DeepInfra at $0.400/M input. For 100M tokens/mo, that is $25/mo saved by routing to Groq.
Sorted by input price
All 5 providers
| Provider | In $/M | Out $/M |
|---|---|---|
| GroqWinner | $0.150 | $0.600 |
 Fireworks AI | $0.180 | $0.720 |
| Cerebras | $0.200 | $0.900 |
 Together AI | $0.220 | $0.600 |
D DeepInfra | $0.400 | $1.20 |
Notes: Groq · Speed leader ; Fireworks AI · Standard serverless ; Cerebras · Wafer-scale chip ; Together AI · Flat pricing ; DeepInfra · Higher markup
Frequently Asked Questions
Groq at $0.150/M input and $0.600/M output. That is 63% cheaper than DeepInfra. Speed leader.