OpenAI: gpt-oss-120b (nitro)
openai/gpt-oss-120b:nitro
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.
Providers
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).
| Provider | Input /M | Output /M | Cache read /M | Latency | Throughput | Uptime |
|---|---|---|---|---|---|---|
| $0.350 | $0.750 | $0.3500 | 0.18s | 703 tok/s | ||
| $0.140 | $0.950 | -- | 1.56s | 350 tok/s | ||
| $0.150 | $0.600 | $0.0750 | 0.25s | 337 tok/s | ||
| $0.150 | $0.600 | -- | 0.43s | 326 tok/s | ||
| $0.150 | $0.600 | $0.1000 | 0.28s | 301 tok/s | ||
| $0.090 | $0.360 | -- | 0.25s | 217 tok/s | ||
| $0.050 | $0.250 | $0.0500 | 0.21s | 185 tok/s | ||
| $0.100 | $0.750 | $0.0550 | 0.47s | 146 tok/s | ||
| $0.100 | $0.500 | $0.1000 | 0.15s | 145 tok/s | ||
| $0.100 | $0.500 | $0.1000 | 0.15s | 143 tok/s | ||
| $0.150 | $0.600 | -- | 0.66s | 131 tok/s | ||
| $0.060 | $0.400 | $0.0300 | 0.83s | 116 tok/s | ||
Fast | $0.150 | $0.750 | -- | 0.27s | 105 tok/s | |
| $0.030 | $0.170 | $0.0300 | 0.22s | 98 tok/s | ||
| $0.030 | $0.187 | $0.0300 | 0.42s | 94 tok/s | ||
| $0.080 | $0.375 | $0.0800 | 0.89s | 94 tok/s | ||
| $0.071 | $0.152 | $0.0640 | 0.58s | 88 tok/s | ||
| $0.150 | $0.600 | -- | 0.25s | 78 tok/s | ||
| $0.030 | $0.150 | $0.0150 | 0.53s | 52 tok/s | ||
| $0.060 | $0.420 | $0.0120 | 0.75s | 49 tok/s | ||
| $0.030 | $0.150 | $0.0300 | 0.68s | 35 tok/s | ||
| $0.150 | $0.600 | $0.0750 | 1.40s | 31 tok/s | ||
| $0.030 | $0.180 | $0.0300 | 3.79s | 14 tok/s | ||
| $0.037 | $0.170 | $0.0185 | -- | -- |
- Input $0.35 / Output $0.75703 tok/s100.00% uptime
- Input $0.14 / Output $0.95350 tok/s99.60% uptime
- Input $0.15 / Output $0.60337 tok/s99.66% uptime
- Input $0.15 / Output $0.60326 tok/s98.36% uptime
- Input $0.15 / Output $0.60301 tok/s99.85% uptime
- Input $0.09 / Output $0.36217 tok/s99.97% uptime
- Input $0.05 / Output $0.25185 tok/s100.00% uptime
- Input $0.10 / Output $0.75146 tok/s99.74% uptime
- Input $0.10 / Output $0.50145 tok/s99.20% uptime
- Input $0.10 / Output $0.50143 tok/s99.65% uptime
- Input $0.15 / Output $0.60131 tok/s98.65% uptime
- Input $0.06 / Output $0.40116 tok/s100.00% uptime
- Input $0.15 / Output $0.75105 tok/s100.00% uptimeFast
- Input $0.03 / Output $0.1798 tok/s96.69% uptime
- Input $0.03 / Output $0.18794 tok/s100.00% uptime
- Input $0.08 / Output $0.37594 tok/s100.00% uptime
- Input $0.071 / Output $0.15288 tok/s100.00% uptime
- Input $0.15 / Output $0.6078 tok/s98.88% uptime
Input $0.03 / Output $0.1552 tok/s99.95% uptime- Input $0.06 / Output $0.4249 tok/s100.00% uptime
- Input $0.03 / Output $0.1535 tok/s99.88% uptime
- Input $0.15 / Output $0.6031 tok/s98.54% uptime
- Input $0.03 / Output $0.1814 tok/s99.34% uptime
Input $0.037 / Output $0.17-- tok/s71.30% uptime
Pricing
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Weighted average
gpt-oss-120b — Price history
| $0.0203 | $0.150 | 64.0% | 6.7% | ||
| $0.0299 | $0.149 | 60.8% | 0.5% | ||
| $0.0299 | $0.169 | 52.5% | 2.0% | ||
| $0.0299 | $0.180 | 19.7% | 1.9% | ||
| $0.0300 | $0.187 | 91.4% | 25.9% | ||
| $0.0383 | $0.399 | 72.1% | 2.8% | ||
| $0.0411 | $0.419 | 38.8% | 0.0% | ||
| $0.0499 | $0.250 | 57.8% | 14.3% | ||
| $0.0767 | $0.750 | 51.6% | 0.4% | ||
| $0.0792 | $0.372 | 34.0% | 0.1% | ||
| $0.0812 | $0.191 | 53.7% | 1.6% | ||
| $0.0899 | $0.360 | 45.7% | 14.0% | ||
| $0.0998 | $0.499 | 54.1% | 1.0% | ||
| $0.0999 | $0.500 | 60.2% | 0.5% | ||
| $0.1088 | $0.599 | 54.9% | 10.5% | ||
| $0.1111 | $0.599 | 77.7% | 0.4% | ||
| $0.1400 | $0.950 | 0.0% | 0.6% | ||
| $0.1494 | $0.599 | 0.0% | 0.0% | ||
| $0.1498 | $0.598 | 36.3% | 0.8% | ||
Fast | $0.1498 | $0.749 | 46.7% | 0.3% | |
| $0.1500 | $0.600 | 0.0% | 2.3% | ||
| $0.1500 | $0.600 | 0.0% | 5.0% | ||
| $0.3499 | $0.749 | 60.9% | 7.2% | ||
| $0.0201 | $0.169 | 91.3% | 1.3% |
- Effective in $0.0203 / Effective out $0.150Token share6.7%
- Effective in $0.0299 / Effective out $0.149Token share0.5%
- Effective in $0.0299 / Effective out $0.169Token share2.0%
- Effective in $0.0299 / Effective out $0.180Token share1.9%
- Effective in $0.0300 / Effective out $0.187Token share25.9%
- Effective in $0.0383 / Effective out $0.399Token share2.8%
- Effective in $0.0411 / Effective out $0.419Token share0.0%
- Effective in $0.0499 / Effective out $0.250Token share14.3%
- Effective in $0.0767 / Effective out $0.750Token share0.4%
- Effective in $0.0792 / Effective out $0.372Token share0.1%
- Effective in $0.0812 / Effective out $0.191Token share1.6%
- Effective in $0.0899 / Effective out $0.360Token share14.0%
- Effective in $0.0998 / Effective out $0.499Token share1.0%
- Effective in $0.0999 / Effective out $0.500Token share0.5%
- Effective in $0.1088 / Effective out $0.599Token share10.5%
- Effective in $0.1111 / Effective out $0.599Token share0.4%
- Effective in $0.1400 / Effective out $0.950Token share0.6%
- Effective in $0.1494 / Effective out $0.599Token share0.0%
- Effective in $0.1498 / Effective out $0.598Token share0.8%
- Effective in $0.1498 / Effective out $0.749FastToken share0.3%
- Effective in $0.1500 / Effective out $0.600Token share2.3%
- Effective in $0.1500 / Effective out $0.600Token share5.0%
- Effective in $0.3499 / Effective out $0.749Token share7.2%
- Effective in $0.0201 / Effective out $0.169Token share1.3%
Performance
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
P50, best across providers
P50, best provider
Uptime
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Availability over the last 3 days
Last 72 hoursAvailability over the last 24 hours
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.
Benchmarks
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | gpt-oss-120b (High) Intelligence Index | 3.8 |
| Artificial Analysis | gpt-oss-120b (High) GPQA Diamond | 78.2% |
| Artificial Analysis | gpt-oss-120b (High) HLE | 19.6% |
| Artificial Analysis | gpt-oss-120b (High) IFBench | 69.0% |
| Artificial Analysis | gpt-oss-120b (High) τ²-Bench Telecom | 65.8% |
| Artificial Analysis | gpt-oss-120b (High) AA-LCR | 52.0% |
| Artificial Analysis | gpt-oss-120b (High) τ³-Banking | 12.8% |
| Artificial Analysis | gpt-oss-120b (High) GDPval-AA | 5.8% |
| Artificial Analysis | gpt-oss-120b (High) CritPt | 1.1% |
| Artificial Analysis | gpt-oss-120b (High) LiveCodeBench | 87.8% |
| Artificial Analysis | gpt-oss-120b (High) SciCode | 34.0% |
| Artificial Analysis | gpt-oss-120b (High) Terminal-Bench Hard | 23.5% |
| Artificial Analysis | gpt-oss-120b (High) Terminal-Bench 4.0 | 0.0% |
| Artificial Analysis | gpt-oss-120b (High) Terminal-Bench 2.1 | 26.2% |
| Artificial Analysis | gpt-oss-120b (High) MMLU-Pro | 80.8% |
| Artificial Analysis | gpt-oss-120b (High) AA-Omniscience Accuracy | 21.7% |
| Artificial Analysis | gpt-oss-120b (High) AA-Omniscience Non-Hallucination Rate | 13.1% |
| Artificial Analysis | gpt-oss-120b (High) AIME 2025 | 93.4% |
| Artificial Analysis | gpt-oss-120b (Low) GPQA Diamond | 67.2% |
| Artificial Analysis | gpt-oss-120b (Low) HLE | 5.9% |
| Artificial Analysis | gpt-oss-120b (Low) IFBench | 58.3% |
| Artificial Analysis | gpt-oss-120b (Low) τ²-Bench Telecom | 45.0% |
| Artificial Analysis | gpt-oss-120b (Low) AA-LCR | 46.0% |
| Artificial Analysis | gpt-oss-120b (Low) τ³-Banking | 2.9% |
| Artificial Analysis | gpt-oss-120b (Low) GDPval-AA | 0.0% |
| Artificial Analysis | gpt-oss-120b (Low) CritPt | 0.0% |
| Artificial Analysis | gpt-oss-120b (Low) LiveCodeBench | 70.7% |
| Artificial Analysis | gpt-oss-120b (Low) Terminal-Bench Hard | 5.3% |
| Artificial Analysis | gpt-oss-120b (Low) Terminal-Bench 2.1 | 13.9% |
| Artificial Analysis | gpt-oss-120b (Low) MMLU-Pro | 77.5% |
| Artificial Analysis | gpt-oss-120b (Low) AIME 2025 | 66.7% |
| Design Arena | GPT OSS 120B Models Arena 3D Elo | 881 |
| Design Arena | GPT OSS 120B Models Arena Code Categories Elo | 951 |
| Design Arena | GPT OSS 120B Models Arena Data Visualization Elo | 984 |
| Design Arena | GPT OSS 120B Models Arena Game Development Elo | 981 |
| Design Arena | GPT OSS 120B Models Arena UI Component Elo | 915 |
| Design Arena | GPT OSS 120B Models Arena Website Elo | 959 |
| Vals AI | GPT OSS 120B AIME | 92.6% |
| Vals AI | GPT OSS 120B CaseLaw v2 | 48.8% |
| Vals AI | GPT OSS 120B CorpFin v2 | 58.2% |
| Vals AI | GPT OSS 120B Finance Agent v1.1 | 21.5% |
| Vals AI | GPT OSS 120B GPQA Diamond | 78.5% |
| Vals AI | GPT OSS 120B LiveCodeBench | 83.2% |
| Vals AI | GPT OSS 120B LegalBench | 75.9% |
| Vals AI | GPT OSS 120B MATH 500 | 94.8% |
| Vals AI | GPT OSS 120B MedQA | 91.4% |
| Vals AI | GPT OSS 120B MGSM | 92.0% |
| Vals AI | GPT OSS 120B MMLU Pro | 79.2% |
| Vals AI | GPT OSS 120B SWE-bench | 33.6% |
| Vals AI | GPT OSS 120B TaxEval v2 | 71.6% |
| Vals AI | GPT OSS 120B Terminal-Bench 2.0 | 19.1% |
Apps
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
Activity
Token volume and request traffic to this model over time.
API
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
Available through EU in-region routing
OpenAI: gpt-oss-120b (nitro) is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).
- Providers in Europe
- Amazon Bedrock
- Prices in Europe
- Amazon Bedrock: $0.15/M tokens input, $0.60/M tokens output
- Base URL
https://eu.openrouter.ai/api/v1- Plan
- Requires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing
Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs
Documentation
Guides, cookbooks, blog posts, and tools for building with OpenAI: gpt-oss-120b (nitro) on OpenRouter.
Frequently asked questions
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 gpt-oss-120b documentation on OpenRouter includes the LLM Quantization: What fp8 and fp4 Do to Quality and Price blog post. The quickstart covers API keys and your first request, and the API reference lists every parameter.
gpt-oss-120b costs $0.03/M input tokens and $0.15/M output tokens, with separate rates for Cache Read at $0.015/M tokens.
gpt-oss-120b has a 131,072 token context window. It supports up to 131,072 completion tokens.
Yes. gpt-oss-120b accepts tools and tool_choice for function calling on 18 of the 21 providers serving it, and requests that send tools are routed to those providers. It also supports structured outputs via a JSON schema in response_format.
gpt-oss-120b is served by 21 providers on OpenRouter: NEAR AI, Venice, CoreWeave, DekaLLM, AkashML, DeepInfra, Prime Intellect, Crusoe and 13 more. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
gpt-oss-120b was released on August 5, 2025. Its knowledge cutoff is June 30, 2024.