Route order log in sign up
1 pick your primary route

OpenAI: gpt-oss-120b

openai/gpt-oss-120b · agent · 131k ctx · added 2025-08-05
$0.1066·0.3741 /M effective hit 47% 64.64B/d

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...

view
openrouter/openai/gpt-oss-120b
OpenAI: gpt-oss-120b · agent · 131k ctx
64.64B/d 47% $0.1066·0.3741 – 1y
sticker
life 429 tok/d dp age
¢ᵉ ¢ⁱ
near-ai/fp4
$0.0300·0.1500·0.0150 $0.0167·0.1500ⁱ $0.0203·0.1489ᵉ 62% – – 8.0% ✓ 550ms 55t/s 1d ·
venice/int4
$0.0300·0.1500·0.0300 $0.0300·0.1500ⁱ $0.0299·0.1488ᵉ 46% – – 0.7% ✓ 889ms 51t/s 1d ·
coreweave/fp4
$0.0300·0.1700·0.0300 $0.0300·0.1700ⁱ $0.0299·0.1690ᵉ 82% – – 6.1% ✓ 234ms 119t/s 3mo ·
dekallm/bf16
$0.0300·0.1800·0.0300 $0.0300·0.1800ⁱ $0.0298·0.1797ᵉ 27% – – 1.1% ✓ 457ms 43t/s 1mo ·
✓
akashml/fp4
$0.0300·0.1870·0.0300 $0.0300·0.1870ⁱ $0.0298·0.1862ᵉ 46% – – 7.1% ✓ 429ms 87t/s 1d ·
deepinfra/bf16
$0.0370·0.1700 $0.0370·0.1700ⁱ $0.0367·0.1688ᵉ 0% – – 6.8% ✓ 411ms 33t/s 3mo ·
≈
primeintellect/fp4
$0.0630·0.1350·0.0570 $0.0577·0.1350ⁱ $0.0518·0.1107ᵉ 39% – – 2.9% ✓ 870ms 63t/s 1d ·
novita/fp4
$0.0500·0.2500 · $0.0500·0.2500ⁱ $0.0499·0.2496ᵉ · 0% · – – – · ✓ 845ms · 89t/s · ✕1mo ·
crusoe/bf16
$0.0500·0.2500·0.0500 $0.0500·0.2500ⁱ $0.0499·0.2494ᵉ 57% – – 10% ✓ 295ms 200t/s 1mo ·
mancer/fp8
$0.0550·0.2500·0.0550 $0.0550·0.2500ⁱ $0.0546·0.2488ᵉ 28% – – <0.1% ✓ 782ms 93t/s 3mo ·
sail-research/fp4
$0.0600·0.4000·0.0300 $0.0335·0.4000ⁱ $0.0395·0.3993ᵉ 64% – – 4.0% ✓ 820ms 77t/s 1d ·
digitalocean
$0.0600·0.4200·0.0120 $0.0176·0.4200ⁱ $0.0525·0.4188ᵉ 18% – – <0.1% ✓ 833ms 45t/s 3mo ·
google-vertex/global
$0.0900·0.3600 $0.0900·0.3600ⁱ $0.0898·0.3592ᵉ 52% – – 7.6% ✓ 190ms 181t/s 3mo ·
baseten/fp4
$0.1000·0.5000·0.1000 $0.1000·0.5000ⁱ $0.0998·0.4991ᵉ 67% – – 0.8% ✓ 137ms 199t/s 3mo ·
baseten/fp4/us
$0.1000·0.5000·0.1000 $0.1000·0.5000ⁱ $0.0998·0.4993ᵉ 58% – – 1.3% ✓ 203ms 212t/s 1d ·
parasail/fp4
$0.1000·0.7500·0.0550 $0.0602·0.7500ⁱ $0.0749·0.7495ᵉ 53% – – 0.6% ✓ 452ms 95t/s 3mo ·
♻
nebius/fp4
$0.1500·0.6000·0.1000 $0.1058·0.6000ⁱ $0.1043·0.5987ᵉ 87% – – 1.5% ✓ 226ms 255t/s 3mo ·
★
groq
$0.1500·0.6000·0.0750 $0.0837·0.6000ⁱ $0.1092·0.5991ᵉ 56% – – 15% ✓ 230ms 305t/s 3mo ·
deepinfra/turbo
$0.1500·0.6000 · $0.1500·0.6000ⁱ $0.1497·0.5988ᵉ · 0% · – – – · ✓ 332ms · 144t/s · ✕1mo ·
amazon-bedrock/eu-west-1
$0.1500·0.6000 $0.1500·0.6000ⁱ $0.1498·0.5988ᵉ 0% – – 8.3% ✓ 421ms 132t/s 3mo ·
siliconflow/fp8
$0.1500·0.6000·0.0750 $0.0837·0.6000ⁱ $0.1498·0.5996ᵉ 0% – – <0.1% ✓ 1113ms 65t/s 3mo ·
phala
$0.1500·0.6000 $0.1500·0.6000ⁱ $0.1499·0.5993ᵉ 1% – – <0.1% ✓ 951ms 115t/s ✕22h ·
together
$0.1500·0.6000 $0.1500·0.6000ⁱ $0.1499·0.5992ᵉ 69% – – 0.4% ✓ 846ms 57t/s 3mo ·
amazon-bedrock
$0.1500·0.6000 $0.1500·0.6000ⁱ $0.1499·0.5995ᵉ 0% – – 3.4% ✓ 589ms 333t/s 3mo ·
parasail/fast
$0.1500·0.7500 $0.1500·0.7500ⁱ $0.1498·0.7494ᵉ 38% – – 0.4% ✓ 313ms 162t/s 1d ·
mara
$0.1500·0.7500 · $0.1500·0.7500ⁱ $0.1499·0.7495ᵉ · 0% · – – – · ✓ 917ms · 197t/s · ✕1mo ·
sambanova
$0.1400·0.9500 $0.1400·0.9500ⁱ $0.1399·0.9494ᵉ 0% – – 0.7% ✓ 863ms 253t/s 3mo ·
deepinfra/fp8
$0.2000·0.9500 $0.2000·0.9500ⁱ $0.1998·0.9487ᵉ 0% – – 0.9% ✓ 1277ms 62t/s 2mo ·
⚡
cerebras/fp16
$0.3500·0.7500·0.3500 $0.3500·0.7500ⁱ $0.3498·0.7491ᵉ 67% – – 12% ✓ 255ms 661t/s 3mo ·
¢
akashml/bf16
$0.0300·0.1700·0.0300 · $0.0300·0.1700ⁱ – · – · – – – · · 538ms · 63t/s · ✕1mo ·
vercel/openai/gpt-oss-120b
GPT OSS 120B · agent · 131k ctx
– 59% $0.1089·0.6000 – 1y
requesty/openai/gpt-oss-120b
gpt-oss-120b · agent · 131k ctx
– $0.1184·0.6000 – 1y
hf/openai/gpt-oss-120b
gpt-oss-120b · agent · 131k ctx
– $0.0950·0.4850 – 1y