Groq vs OpenRouter
Same workloads, both price lists, refreshed daily. On shared line items today: Groq is cheaper on 2 of 7 shared models (input or output price), OpenRouter on 4.
Groq
Custom LPU silicon serving open models at hundreds of tokens/second: the speed king for latency-sensitive apps, with simple per-token pricing..
- Fastest tokens/sec in the market
- Simple pricing
- Generous free tier for prototyping
OpenRouter
One API key for 340+ models across every major provider, with automatic failover and pass-through pricing. The fastest way to hedge provider risk or A/B models without new integrations.
- 340+ models, one integration
- Automatic provider failover
- Pass-through pricing, small platform fee
Shared models, priced by both
List prices refreshed 2026-08-23 ยท input $/1M tokens| Model | Groq in/out | OpenRouter in/out | Cheaper input |
|---|---|---|---|
| GPT-OSS 20B | $0.075 / $0.300 | $0.020 / $0.100 | OpenRouter |
| GPT-OSS-Safeguard-20b | $0.075 / $0.300 | $0.075 / $0.300 | tie |
| GPT-OSS 120B | $0.150 / $0.600 | $0.180 / $0.800 | Groq |
| Llama Guard 4 12B | $0.200 / $0.200 | $0.180 / $0.180 | OpenRouter |
| Qwen3 32B | $0.290 / $0.590 | $0.080 / $0.280 | OpenRouter |
| Qwen3.6 27B | $0.600 / $3.00 | $0.600 / $3.60 | tie |
| Kimi K2 | $1.00 / $3.00 | $0.570 / $2.30 | OpenRouter |
Cheaper-on-count is a tally of listed prices, not a quality verdict: throughput, reliability and quantization differ between providers. Disclosure.