gpt-oss-120b pricing.
prices as of · 68 providers · 83 prices
OpenAI's gpt-oss-120b holds 131k tokens of context and has a reasoning mode. Open weights, so anyone may host it, which is why the prices below differ so much.
gpt-oss-120b costs $0.04 in and $0.17 out per million tokens from OpenRouter; the cheapest price we can explain is $0.24 at LLM Gateway, checked 16 Sept 2026.
OpenRouter list price
$0.28/M tokens
$0.04 in · $0.17 out
read · 92% below the market median
Cheapest price we can explain
$0.24/M tokens
$0.03 in · $0.14 out · 16% under list
LLM Gatewayread
What it costs for your work
OpenRouter list price
$0.0324a month
cheapest we can explain · LLM Gateway
$0.0276a month
difference
$0.0048saved
A page is about 600 tokens in and 60 out, a support reply 200 in and 300 out, an email 150 in and 250 out. Nothing else is counted: no cache discount, no batch rate, no free tier.
Run it yourself
every plan that fits gpt-oss-120b →OpenAI publishes the weights, so you can run gpt-oss-120b on a server you rent by the month. It needs about 65.5 GB of RAM on a CPU, and the cheapest plan we track with that much and a disk included is Linode Linode 90GB at $240/mo, streaming an estimated 2.5–5.1 tok/s.
Every provider selling gpt-oss-120b
Four groups, decided by the price and by what the provider publishes about it. Prices are per million tokens, and the share is that row against OpenRouter's price. OpenAI trained gpt-oss-120b and sells it on no catalog we read, so the price every row is measured against is OpenRouter's.
Cheaper, and we know why
10 pricesBelow list, with a published reason: hosting the open weights, a regional price list, credits, a stated discount.
$0.03 / $0.14in / out per M
0.84×of list price
$0.03 / $0.14in / out per M
0.84×of list price
OrcaRouterHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$0.03 / $0.17in / out per M
0.93×of list price
Weights & BiasesHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
$0.03 / $0.17in / out per M
0.93×of list price
$0.03 / $0.17in / out per M
0.93×of list price
AkashMLHosts the open weights on its own servers, so the price is its own, not a resale. Precision bf16.
bf16 precision$0.03 / $0.17in / out per M
0.93×of list price
CoreWeaveServes the open weights at fp4, a lower precision than the maker's, so not quite the same product.
fp4 precision$0.03 / $0.17in / out per M
0.93×of list price
Nous PortalResells OpenRouter's catalog at 0.80× the listed price, funded by a subscription whose credits carry a 10% bonus.
$0.03 / $0.17in / out per M
0.93×of list price
DekaLLMHosts the open weights on its own servers, so the price is its own, not a resale. Precision bf16.
bf16 precision$0.03 / $0.18in / out per M
0.96×of list price
$0.03 / $0.18in / out per M
0.98×of list price
Same price as the maker
65 pricesThe same price the maker charges, sold by someone else.
Helicone8% above OpenRouter's price.
$0.04 / $0.16in / out per M
1.00×of list price
$0.04 / $0.17in / out per M
1.00×of list price
$0.04 / $0.17in / out per M
1.00×of list price
$0.04 / $0.17in / out per M
1.00×of list price
$0.04 / $0.17in / out per M
1.00×of list price
- TensorX
8% above OpenRouter's price.
$0.04 / $0.20in / out per M
1.14×of list price
$0.05 / $0.20in / out per M
1.25×of list price
$0.05 / $0.25in / out per M
1.42×of list price
$0.05 / $0.25in / out per M
1.42×of list price
$0.05 / $0.25in / out per M
1.42×of list price
Synthetic170% above OpenRouter's price.
$0.10 / $0.10in / out per M
1.42×of list price
DInference82% above OpenRouter's price.
$0.07 / $0.27in / out per M
1.68×of list price
$0.06 / $0.42in / out per M
2.14×of list price
$0.05 / $0.45in / out per M
2.14×of list price
$0.09 / $0.36in / out per M
2.24×of list price
Merge Gateway143% above OpenRouter's price.
$0.09 / $0.36in / out per M
2.24×of list price
$0.09 / $0.36in / out per M
2.24×of list price
$0.08 / $0.44in / out per M
2.42×of list price
$0.10 / $0.40in / out per M
2.49×of list price
Cortecs149% above OpenRouter's price.Routes only to EU-hosted providers, prices in euros, and adds a flat 5% fee.
131k context hereCortecs ↗$0.09 / $0.46in / out per M
2.63×of list price
OpenReason185% above OpenRouter's price.
$0.11 / $0.42in / out per M
2.63×of list price
$0.09 / $0.47in / out per M
2.63×of list price
$0.10 / $0.50in / out per M
2.85×of list price
$0.10 / $0.50in / out per M
2.85×of list price
- Submodel
170% above OpenRouter's price.
$0.10 / $0.50in / out per M
2.85×of list price
showing the 25 cheapest of 65 · every row in the open dataset
Cheaper, and nobody says why
6 prices · never rankedBelow list on a closed model with nothing to account for it. Shown, never ranked, never a pick.
KenariWe could not read this provider's prices reliably, so the row is listed but never ranked.
$0.06 / $0.35in / out per M
1.92×of list price
MancerWe could not read this provider's prices reliably, so the row is listed but never ranked.
fp8 precisionMancer ↗$0.05 / $0.40in / out per M
1.96×of list price
TinfoilWe could not read this provider's prices reliably, so the row is listed but never ranked.
$0.15 / $0.60in / out per M
3.74×of list price
FeatherlessWe could not read this provider's prices reliably, so the row is listed but never ranked.
$0.15 / $0.60in / out per M
3.74×of list price
SambaNovaWe could not read this provider's prices reliably, so the row is listed but never ranked.
$0.22 / $0.59in / out per M
4.45×of list price
- Mancer
We could not read this provider's prices reliably, so the row is listed but never ranked.
$20,000 / $160,000in / out per M
782918.15×of list price
Given away with limits
2 free tiers · never rankedThese providers give gpt-oss-120b away inside a quota of their own. A limit is not a price, so none of them counts as cheapest here or anywhere else on the site.
NVIDIA build- QVAC
Frequently asked
- Who sells gpt-oss-120b cheapest?
- LLM Gateway, at $0.03 in and $0.14 out per million tokens, read on 16 Sept 2026. Routes to a host serving the open weights; precision not disclosed.
- Is gpt-oss-120b open weights?
- Yes. OpenAI publishes the weights, so anyone may host gpt-oss-120b, which is why 68 providers sell it and why their prices differ so much.
- How much context does gpt-oss-120b hold?
- 131,072 tokens in one call, which is about 131k, and it can call your own functions.
- How does gpt-oss-120b compare with the rest of the market?
- At $0.28 per million tokens, OpenRouter's price is 92% below the $3.55 median across the 337 text models we track.