GPT OSS 120B

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 to run on a single H100 GPU with native MXFP4 quantization. The model supports configurable reasoning depth, full chain-of-thought access, and native tool use, including function calling, browsing, and structured output generation.

openai/gpt-oss-120b
Organization
OpenAI
Family
gpt-oss
Providers
39
Context
131,072
Output limit
32,768
Knowledge
—
Release
2025-08-05
Updated
2025-08-05
Weights
Open
Input
text
Output
text
Capabilities
Tools, Reasoning, Structured, Temperature

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

87 offers · USD per million tokens
ProviderModel IDContextOutputInput / output · 1MReasoningToolsStructuredDetails
Abacus
openai/gpt-oss-120b
128,00016,384$0.08 / $0.44YesYesYes
View
ai&
openai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
AKI.IO
gpt-oss-120b
128,00032,768$0.15 / $0.55YesYesYes
View
Amazon Bedrock
us-gov.openai.gpt-oss-120b-1:0
128,00016,384$0.18 / $0.72YesYesYes
View
Amazon Bedrock
openai.gpt-oss-120b
131,072131,072$0.15 / $0.60YesYesYes
View
Amazon Bedrock
openai.gpt-oss-120b-1:0
131,072128,000$0.15 / $0.60YesYesYes
View
Cerebras
gpt-oss-120b
131,07240,960$0.35 / $0.75YesYesYes
View
Charm Hyper
gpt-oss-120b
131,07213,107$0.178 / $0.68YesYesYes
View
CloudFerro Sherlock
openai/gpt-oss-120b
131,000131,000$2.92 / $2.92YesYesYes
View
Cloudflare Workers AI
@cf/openai/gpt-oss-120b
128,00016,384$0.35 / $0.75YesYesYes
View
CoreWeave
openai/gpt-oss-120b
131,072131,072$0.03 / $0.17YesYesYes
View
Cortecs
gpt-oss-120b
131,000131,000$0.089 / $0.446YesYesYes
View
Crusoe
openai/gpt-oss-120b
131,07232,768$0.05 / $0.20YesYesYes
View
Eden AI
together_ai/openai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Eden AI
nebius/openai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Eden AI
cerebras/gpt-oss-120b
131,07232,768$0.35 / $0.75YesYesYes
View
Eden AI
ionos/openai/gpt-oss-120b
131,07232,768$0.168375 / $0.729625YesNoNo
View
Eden AI
flexai/gpt-oss-120b
131,07232,768$0.03 / $0.17YesYesYes
View
Eden AI
ovhcloud/gpt-oss-120b
131,07232,768$0.09 / $0.47YesNoYes
View
Eden AI
cloudflare/@cf/openai/gpt-oss-120b
128,00032,768$0.35 / $0.75YesYesNo
View
Eden AI
deepinfra/openai/gpt-oss-120b
131,07232,768$0.037 / $0.17YesYesYes
View
Eden AI
fireworks_ai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesNo
View
Eden AI
databricks/databricks-gpt-oss-120b@eu
131,07232,768$0.15001 / $0.59997YesYesNo
View
Eden AI
scaleway/gpt-oss-120b
128,00032,768$0.168375 / $0.6735YesYesNo
View
Eden AI
groq/openai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Eden AI
fireworks_ai/accounts/fireworks/models/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Eden AI
databricks/databricks-gpt-oss-120b
131,07232,768$0.15001 / $0.59997YesYesNo
View
GreenPT
gpt-oss-120b
131,07232,768$0.228 / $0.798YesYesYes
View
GreenPT
green-r-raw
131,07232,768$0.399 / $1.083YesYesYes
View
GreenPT
green-r
131,07232,768$0.399 / $1.083YesYesYes
View
Groq
openai/gpt-oss-120b
131,07265,536$0.15 / $0.60YesYesYes
View
Hugging Face
openai/gpt-oss-120b
131,07232,768$0.25 / $0.69YesYesYes
View
Impossibl
cerebras/gpt-oss-120b
131,07232,768$0.35 / $0.75YesYesYes
View
Impossibl
fireworks/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Impossibl
groq/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Kenari
gpt-oss-120b
131,07232,768$0.00 / $0.00YesYesYes
View
Kilo Gateway
openai/gpt-oss-120b
131,072117,964$0.03 / $0.17YesYesYes
View
LLM Gateway
scx-ai/gpt-oss-120b
131,07232,768$0.17 / $0.55YesYesNo
View
LLM Gateway
bytedance/gpt-oss-120b
128,00032,000$0.10 / $0.50YesYesNo
View
LLM Gateway
groq/gpt-oss-120b
131,07232,766$0.15 / $0.75YesYesNo
View
LLM Gateway
runware/gpt-oss-120b
131,07232,768$0.032 / $0.14YesYesNo
View
LLM Gateway
cerebras/gpt-oss-120b
131,07232,768$0.35 / $0.75YesYesNo
View
LLM Gateway
azure/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesNo
View
LLM Gateway
together-ai/gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesNo
View
Merge Gateway
openai/gpt-oss-120b
128,00032,000$0.05 / $0.25YesNoNo
View
NanoGPT
openai/gpt-oss-120b
128,00016,384$0.35 / $0.75YesYesYes
View
NanoGPT
TEE/gpt-oss-120b
131,07216,384$2.00 / $2.00NoNoNo
View
Neon
gpt-oss-120b
131,07225,000$0.15 / $0.60YesYesYes
View
OCI Generative AI
openai.gpt-oss-120b
128,00016,384$0.15 / $0.60YesYesYes
View
OpenReason
openai/gpt-oss-120b
131,07232,768$0.1055 / $0.422YesYesYes
View
AkashMLvia OpenRouter
openai/gpt-oss-120b
bf16
131,072117,964$0.037 / $0.187YesYesYes
View
Amazon Bedrockvia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.15 / $0.60YesNoNo
View
Amazon Bedrockvia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.15 / $0.60YesNoNo
View
BaseTenvia OpenRouter
openai/gpt-oss-120b
fp4
128,072115,264$0.10 / $0.50YesYesYes
View
Cerebrasvia OpenRouter
openai/gpt-oss-120b
fp16
131,07240,960$0.35 / $0.75YesYesYes
View
CoreWeavevia OpenRouter
openai/gpt-oss-120b
fp4
131,072117,964$0.03 / $0.17YesYesYes
View
Crusoevia OpenRouter
openai/gpt-oss-120b
bf16
131,072117,964$0.05 / $0.25YesYesYes
View
DeepInfravia OpenRouter
openai/gpt-oss-120b
bf16
131,07216,384$0.15 / $0.60YesYesYes
View
DeepInfravia OpenRouter
openai/gpt-oss-120b
bf16
131,072117,964$0.037 / $0.17YesYesYes
View
DekaLLMvia OpenRouter
openai/gpt-oss-120b
bf16
131,072117,964$0.03 / $0.18YesYesYes
View
DigitalOceanvia OpenRouter
openai/gpt-oss-120b
unknown
128,0004,096$0.06 / $0.42YesNoNo
View
Googlevia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.09 / $0.36YesNoYes
View
Groqvia OpenRouter
openai/gpt-oss-120b
unknown
131,07265,536$0.15 / $0.60YesYesYes
View
Mancer 2via OpenRouter
openai/gpt-oss-120b
fp8
131,072117,964$0.045 / $0.275YesYesYes
View
Maravia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.15 / $0.75YesYesYes
View
Nebiusvia OpenRouter
openai/gpt-oss-120b
fp4
131,072117,964$0.15 / $0.60YesYesYes
View
Novitavia OpenRouter
openai/gpt-oss-120b
fp4
131,07232,768$0.05 / $0.25YesYesNo
View
Parasailvia OpenRouter
openai/gpt-oss-120b
fp4
131,072117,964$0.10 / $0.75YesYesYes
View
Phalavia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.15 / $0.60YesYesYes
View
SambaNovavia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.14 / $0.95YesYesNo
View
SiliconFlowvia OpenRouter
openai/gpt-oss-120b
fp8
131,0728,192$0.15 / $0.60YesNoYes
View
Togethervia OpenRouter
openai/gpt-oss-120b
unknown
131,072117,964$0.15 / $0.60YesYesYes
View
Opper
gpt-oss-120b
128,0008,192$1.1622 / $4.88124YesYesYes
View
OrcaRouter
openai/gpt-oss-120b
131,07232,768$0.03 / $0.17YesYesYes
View
Pendra
gpt-oss:120b
131,07232,768$0.00 / $0.00YesYesYes
View
Pioneer
openai/gpt-oss-120b
131,072131,072$0.15 / $0.60YesYesYes
View
Privatemode AI
gpt-oss-120b
128,00032,768$0.2311 / $0.7511YesYesYes
View
QVAC
gpt-oss-120b
131,07232,768$0.00 / $0.00YesYesYes
View
SCX.ai
gpt-oss-120b
131,072131,072$0.17 / $0.55YesYesYes
View
STACKIT
openai/gpt-oss-120b
131,0008,192$0.53 / $0.76YesYesNo
View
Tempr Gateway
cerebras/gpt-oss-120b
131,07240,960$0.35 / $0.75YesYesYes
View
Tempr Gateway
groq/openai/gpt-oss-120b
131,07265,536$0.15 / $0.60YesYesYes
View
Tinfoil
gpt-oss-120b
131,07232,768$0.15 / $0.60YesYesYes
View
Venice AI
openai-gpt-oss-120b
128,00016,384$0.07 / $0.30YesYesYes
View
Vercel AI Gateway
openai/gpt-oss-120b
131,072131,072$0.10 / $0.50YesYesYes
View
Vertex
openai/gpt-oss-120b-maas
131,072131,072$0.09 / $0.36YesYesYes
View
watsonx.ai
openai/gpt-oss-120b
131,07232,768$0.159 / $0.636YesYesYes
View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens
Input from$0.00Per 1M input tokens
Output from$0.00Per 1M output tokens
Available offers87Across 39 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 10 lowest-priced configurations
InputOutput
CoreWeavecoreweave/fp4
$0.03$0.17
DekaLLMdekallm/bf16
$0.03$0.18
AkashMLakashml/bf16
$0.037$0.187
DeepInfradeepinfra/bf16
$0.037$0.17
Mancer 2mancer/fp8
$0.045$0.275
Crusoecrusoe/bf16
$0.05$0.25
Novitanovita/fp4
$0.05$0.25
DigitalOceandigitalocean
$0.06$0.42
Googlegoogle-vertex/global
$0.09$0.36
BaseTenbaseten/fp4
$0.10$0.50
openai/gpt-oss-120bObserved
Input
$0.037
Output
$0.17
Hosting prices through OpenRouter All endpoints and cache rates
EndpointInput / 1MOutput / 1MCache read / 1MCache write / 1M
AkashMLakashml/bf16$0.037$0.187$0.037—
Amazon Bedrockamazon-bedrock$0.15$0.60——
Amazon Bedrockamazon-bedrock/eu-west-1$0.15$0.60——
BaseTenbaseten/fp4$0.10$0.50$0.10—
Cerebrascerebras/fp16$0.35$0.75$0.35—
CoreWeavecoreweave/fp4$0.03$0.17$0.03—
Crusoecrusoe/bf16$0.05$0.25$0.05—
DeepInfradeepinfra/turbo$0.15$0.60——
DeepInfradeepinfra/bf16$0.037$0.17——
DekaLLMdekallm/bf16$0.03$0.18$0.03—
DigitalOceandigitalocean$0.06$0.42$0.012—
Googlegoogle-vertex/global$0.09$0.36——
Groqgroq$0.15$0.60$0.075—
Mancer 2mancer/fp8$0.045$0.275——
Maramara$0.15$0.75——
Nebiusnebius/fp4$0.15$0.60——
Novitanovita/fp4$0.05$0.25——
Parasailparasail/fp4$0.10$0.75$0.055—
Phalaphala$0.15$0.60——
SambaNovasambanova$0.14$0.95——
SiliconFlowsiliconflow/fp8$0.15$0.60$0.075—
Togethertogether$0.15$0.60——

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 6 endpoints
P50P75P90P99
AkashMLakashml/bf16
441–6,851.18 ms
Amazon Bedrockamazon-bedrock
518–4,884.7 ms
Cerebrascerebras/fp16
192–1,127.45 ms
CoreWeavecoreweave/fp4
418–10,404.35 ms
Crusoecrusoe/bf16
297–1,130.14 ms
DeepInfradeepinfra/turbo
464–1,880.06 ms
All latency percentiles 20 endpoints
EndpointP50P75P90P99
AkashMLakashml/bf16441 ms590 ms1,290.7 ms6,851.18 ms
Amazon Bedrockamazon-bedrock518 ms646 ms949 ms4,884.7 ms
Cerebrascerebras/fp16192 ms327 ms452 ms1,127.45 ms
CoreWeavecoreweave/fp4418 ms765 ms3,118.8 ms10,404.35 ms
Crusoecrusoe/bf16297 ms346 ms437 ms1,130.14 ms
DeepInfradeepinfra/turbo464 ms565 ms811 ms1,880.06 ms
DeepInfradeepinfra/bf16376 ms522 ms1,293.8 ms4,610.72 ms
DekaLLMdekallm/bf16815 ms1,150 ms2,881.8 ms15,388.17 ms
DigitalOceandigitalocean545.5 ms900 ms1,512.9 ms3,225.7 ms
Googlegoogle-vertex/global1,290 ms5,441 ms9,141.4 ms114,124.16 ms
Groqgroq141 ms227 ms421 ms1,841.27 ms
Mancer 2mancer/fp8886 ms1,075 ms1,256 ms3,164.45 ms
Maramara980 ms1,162.5 ms1,675.4 ms3,499.82 ms
Nebiusnebius/fp4253.5 ms366 ms582.8 ms1,300.27 ms
Novitanovita/fp4776 ms981 ms1,408.8 ms3,326.56 ms
Parasailparasail/fp4642 ms804 ms967.9 ms1,546.09 ms
Phalaphala909 ms1,104 ms1,683 ms4,782.24 ms
SambaNovasambanova964 ms1,402.75 ms2,314.5 ms5,790.98 ms
SiliconFlowsiliconflow/fp81,706 ms2,534 ms3,539.5 ms8,084.2 ms
Togethertogether208 ms319.5 ms502 ms1,112.8 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 6 endpoints
P50P75P90P99
AkashMLakashml/bf16
110–201 t/s
Amazon Bedrockamazon-bedrock
284–497.42 t/s
Cerebrascerebras/fp16
873–1,945 t/s
CoreWeavecoreweave/fp4
34–65 t/s
Crusoecrusoe/bf16
231–287.57 t/s
DeepInfradeepinfra/turbo
115–548 t/s
All throughput percentiles 20 endpoints
EndpointP50P75P90P99
AkashMLakashml/bf16110 t/s133 t/s157 t/s201 t/s
Amazon Bedrockamazon-bedrock284 t/s360 t/s417 t/s497.42 t/s
Cerebrascerebras/fp16873 t/s1,158 t/s1,399 t/s1,945 t/s
CoreWeavecoreweave/fp434 t/s39 t/s47 t/s65 t/s
Crusoecrusoe/bf16231 t/s251 t/s265 t/s287.57 t/s
DeepInfradeepinfra/turbo115 t/s166 t/s288 t/s548 t/s
DeepInfradeepinfra/bf1662 t/s70 t/s79 t/s95 t/s
DekaLLMdekallm/bf1618 t/s21 t/s24 t/s30 t/s
DigitalOceandigitalocean66 t/s72 t/s76 t/s80 t/s
Googlegoogle-vertex/global74 t/s204 t/s269 t/s389.21 t/s
Groqgroq352 t/s403 t/s435 t/s517.09 t/s
Mancer 2mancer/fp898 t/s131 t/s166 t/s266.09 t/s
Maramara202 t/s243 t/s301.1 t/s536.61 t/s
Nebiusnebius/fp4254 t/s302 t/s344 t/s420 t/s
Novitanovita/fp4152 t/s178.25 t/s201.5 t/s241.25 t/s
Parasailparasail/fp4100 t/s120 t/s160 t/s264 t/s
Phalaphala136 t/s169 t/s196 t/s273.86 t/s
SambaNovasambanova375.5 t/s473.75 t/s529 t/s647.45 t/s
SiliconFlowsiliconflow/fp834 t/s46 t/s69.1 t/s115.83 t/s
Togethertogether140 t/s161 t/s184 t/s217.06 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window
Availability by endpointCurrent overlapping windows · 22 endpoints
Endpoint5 minutes30 minutes24 hours
AkashMLakashml/bf16
Amazon Bedrockamazon-bedrock
Amazon Bedrockamazon-bedrock/eu-west-1
BaseTenbaseten/fp4
Cerebrascerebras/fp16
CoreWeavecoreweave/fp4
Crusoecrusoe/bf16
DeepInfradeepinfra/turbo
DeepInfradeepinfra/bf16
DekaLLMdekallm/bf16
DigitalOceandigitalocean
Googlegoogle-vertex/global
Groqgroq
Mancer 2mancer/fp8
Maramara
Nebiusnebius/fp4
Novitanovita/fp4
Parasailparasail/fp4
Phalaphala
SambaNovasambanova
SiliconFlowsiliconflow/fp8
Togethertogether

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations
openai/gpt-oss-120bObserved
Benchmark profileArtificial Analysis · shared source scale from 0 to 40
Coding30.4
Agentic3.7
Intelligence11.6
Design ArenaCategoryEloWin rateRank
models3d91629.4%111
modelscodecategories96933.4%120
modelsdataviz99543.6%110
modelsgamedev1,00540.5%110
modelsuicomponent93435.7%116
modelswebsite97432.5%122

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better
tau bench verified airlineOpenAI: gpt-oss-120b
63.417%
gpqa diamondOpenAI: gpt-oss-120b
72.559%
EvaluationConfigurationSourceResultsDetails
models · websiteGPT OSS 120Bopenai/gpt-oss-120bDesign Arena974 Elo
Details
models · gamedevGPT OSS 120Bopenai/gpt-oss-120bDesign Arena1,005 Elo
Details
tau bench verified airlineOpenAI: gpt-oss-120bopenai/gpt-oss-120bOpenRouter63.417%
Details
models · uicomponentGPT OSS 120Bopenai/gpt-oss-120bDesign Arena934 Elo
Details
models · 3dGPT OSS 120Bopenai/gpt-oss-120bDesign Arena916 Elo
Details
Composite indicesgpt-oss-120b (high)openai/gpt-oss-120bArtificial AnalysisIntelligence 11.6 · Coding 30.4 · Agentic 3.7
Details
gpqa diamondOpenAI: gpt-oss-120bopenai/gpt-oss-120bOpenRouter72.559%
Details
models · codecategoriesGPT OSS 120Bopenai/gpt-oss-120bDesign Arena969 Elo
Details
models · datavizGPT OSS 120Bopenai/gpt-oss-120bDesign Arena995 Elo
Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 4, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Sep 27, 2026
ApplicationModel routeTurnsMedian cost / sessionDetails
Claude Codeopenai/gpt-oss-120b2–9 turns
$0.00282
Details
Hermes Agentopenai/gpt-oss-120b1 turn
$0.000484
Details
Hermes Agentopenai/gpt-oss-120b10–49 turns
$0.022736
Details
Hermes Agentopenai/gpt-oss-120b2–9 turns
$0.002069
Details

Source: OpenRouter, as of Oct 2, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Activity

Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.

Last 90 complete UTC days
Daily token usage6,593,970,728,503 reported tokens · 89 days with observations
Jul 6, 2026Oct 3, 2026
Daily observations · 89 days
Date (UTC)Reported tokensReported routes
Oct 3, 202656,082,356,4831
Oct 2, 202676,460,953,4901
Oct 1, 202672,030,012,3401
Sep 30, 2026104,414,061,1301
Sep 29, 2026108,027,521,5121
Sep 28, 202679,581,300,0281
Sep 27, 202655,279,280,0281
Sep 26, 202673,412,190,0311
Sep 25, 2026100,226,860,6561
Sep 24, 2026112,082,050,5291
Sep 23, 202699,911,502,7041
Sep 22, 2026104,632,368,8051
Sep 21, 202685,947,899,8181
Sep 20, 202658,608,187,0321
Sep 19, 202653,710,725,5111
Sep 18, 202667,376,759,8061
Sep 17, 202681,542,601,9331
Sep 16, 2026150,784,489,6581
Sep 15, 2026127,780,122,0971
Sep 14, 202683,933,730,3371
Sep 13, 202656,186,586,8581
Sep 12, 202672,299,306,9631
Sep 11, 2026112,468,461,1721
Sep 10, 202692,430,113,8661
Sep 9, 202683,465,059,1081
Sep 8, 202674,491,045,9981
Sep 7, 202657,891,749,1131
Sep 6, 202651,293,909,8951
Sep 5, 202647,150,586,4131
Sep 4, 202663,984,868,9331
Sep 3, 202663,106,485,8801
Sep 2, 202681,970,720,1211
Sep 1, 202689,596,375,0441
Aug 31, 202664,869,533,3741
Aug 30, 202637,999,951,6261
Aug 29, 202635,615,007,1201
Aug 28, 202676,449,207,2551
Aug 27, 202682,893,119,8801
Aug 26, 2026116,194,111,0881
Aug 25, 202665,330,217,9241
Aug 24, 202657,694,468,9061
Aug 23, 202654,396,515,9591
Aug 22, 202672,004,265,2701
Aug 21, 202669,689,203,4931
Aug 20, 202685,473,416,0591
Aug 19, 202669,087,964,4021
Aug 18, 202689,378,420,9961
Aug 17, 202676,270,529,2221
Aug 16, 202650,293,655,3881
Aug 15, 202650,508,503,5421
Aug 14, 202665,009,932,7211
Aug 13, 202667,713,330,4551
Aug 12, 202671,511,635,4841
Aug 11, 202676,683,977,3691
Aug 10, 202661,296,259,3301
Aug 9, 202657,717,627,8821
Aug 8, 202648,857,258,2241
Aug 7, 202652,892,303,3881
Aug 6, 202652,987,465,2641
Aug 5, 202659,116,643,9621
Aug 4, 202680,968,943,4991
Aug 3, 202665,741,515,1791
Aug 2, 202669,855,250,0391
Aug 1, 202676,986,037,9861
Jul 31, 202678,186,210,2711
Jul 30, 202680,598,169,8971
Jul 29, 202676,430,452,6101
Jul 28, 202686,942,273,0731
Jul 27, 202673,626,212,2911
Jul 26, 202665,512,114,5681
Jul 25, 202659,725,595,9991
Jul 24, 202662,517,068,6031
Jul 23, 202674,810,625,3511
Jul 22, 202672,432,675,6141
Jul 21, 202666,650,452,4511
Jul 20, 202668,334,208,3001
Jul 19, 202656,882,848,8141
Jul 18, 202661,998,703,4021
Jul 17, 202671,719,178,3041
Jul 16, 202690,363,003,1321
Jul 15, 202693,945,024,7601
Jul 14, 202681,888,522,5681
Jul 13, 202671,985,891,8891
Jul 12, 202663,832,184,6371
Jul 11, 202666,771,050,8391
Jul 10, 202669,229,151,3041
Jul 9, 202680,844,483,0821
Jul 8, 202683,934,213,4431
Jul 7, 202675,163,897,7231

Source: OpenRouter (openrouter.ai/rankings), as of Oct 4, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications
openai/gpt-oss-120bObserved
Context
131,072
Maximum output
117,964
Tokenizer
GPT
Input
text
Output
text
Moderation
No
Knowledge cutoff
2024-06-30
Added to OpenRouter
2025-08-05
Base URLhttps://openrouter.ai/api/v1
Model IDopenai/gpt-oss-120b
Version IDopenai/gpt-oss-120b
Weights IDopenai/gpt-oss-120b

Reasoning

Required
Yes
Enabled by default
Not reported
Default effort
medium
Supported efforts
high, medium, low

Supported parameters

  • frequency_penalty
  • include_reasoning
  • logit_bias
  • logprobs
  • max_tokens
  • min_p
  • presence_penalty
  • reasoning
  • reasoning_effort
  • repetition_penalty
  • response_format
  • seed
  • stop
  • structured_outputs
  • temperature
  • tool_choice
  • tools
  • top_a
  • top_k
  • top_logprobs
  • top_p

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does GPT OSS 120B cost?

Listed rates start at $0.00 per million input tokens and $0.00 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer GPT OSS 120B?

There are 87 cataloged offers across 39 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 131,072 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Yes. Structured output: Yes. Support can vary by endpoint.

Resources

Catalog history · 2 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

  • · 565ae3e125e5
  • · 80d28654c7e0