
OpenAI · Chat / LLM · 121.7B Parameters · 256K Context

Function Calling Tool Calling Streaming Reasoning Agent Workflows Long Context CodeOverview
Introducing gpt-oss-120b, OpenAI’s flagship open-weight model in the gpt-oss series, built for advanced reasoning, large-scale agentic workloads, and enterprise-grade automation. With 120B parameters and a highly optimized Mixture-of-Experts (MoE) architecture, it activates 12B parameters during inference, delivering exceptional intelligence while maintaining competitive latency. Designed for complex reasoning, multi-task agents, and long-horizon planning, gpt-oss-120b brings frontier-level capability to commercial and self-hosted deployments.Model Specifications
| Field | Details |
|---|---|
| Model ID | openai/gpt-oss-120b |
| Provider | OpenAI |
| Kind | Chat / LLM |
| Architecture | Large-Scale Mixture-of-Experts (MoE) with adaptive routing, SwiGLU activations, hierarchical sparse attention, and token-choice MoE for reasoning efficiency |
| Model Size | 121.7B Params |
| Context Length | 256K Tokens |
| MoE | No |
| Release Date | August 2024 |
| License | Apache 2.0 |
| Training Data | Extensive multi-domain knowledge corpus with safety-aligned fine-tuning, enterprise & community feedback loops, and agentic task simulation datasets |
| Function Calling | Supported |
| Serverless API | Available |
| Fine-tuning | Coming Soon |
| On-demand | Coming Soon |
Pricing
Access via Qubrid’s serverless API with pay-per-token pricing. No infrastructure management required.
| Token Type | Price per 1M Tokens |
|---|---|
| Input Tokens | $0.15 |
| Output Tokens | $0.61 |
Quickstart
Prerequisites
- Create a free account at platform.qubrid.com
- Generate your API key from the API Keys section
- Replace
QUBRID_API_KEYin the code below with your actual key
Python
JavaScript
Go
cURL
Live Example
Prompt: Explain quantum computing in simple terms
Response:
Playground Features
The Qubrid Playground supports advanced prompt engineering features out of the box:🧠 System Prompt
Set a persistent instruction that shapes how the model behaves across the entire conversation.🎯 Few-Shot Examples
Guide the model by showing it example input/output pairs before your actual query — no fine-tuning needed.| User Input | Assistant Response |
|---|---|
What is a closure in JS? | A closure is a function that retains access to its outer scope even after the outer function has returned... |
Explain recursion | Recursion is when a function calls itself. Base case stops the loop. Example: factorial(n) = n * factorial(n-1) |
💡 Few-shot examples are powerful for domain-specific formatting, tone control, and structured outputs — available directly in the playground UI.
Inference Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| Streaming | boolean | true | Enable streaming responses for real-time output. |
| Temperature | number | 0.7 | Controls randomness. Higher values mean more creative but less predictable output. |
| Max Tokens | number | 4096 | Maximum number of tokens to generate in the response. |
| Top P | number | 1 | Nucleus sampling: considers tokens with top_p probability mass. |
| Reasoning Effort | select | medium | Controls how much reasoning effort the model should apply. |
| Reasoning Summary | select | concise | Controls the level of explanation in the reasoning summary. |
Use Cases
- Autonomous agents and multi-step reasoning
- Advanced function calling and workflow orchestration
- Research-grade problem solving and planning
- Enterprise automation across verticals
- Large-scale code generation and debugging
- R&D assistance and scientific exploration
- Conversational AI and smart copilots
- Knowledge extraction and document understanding
- Long-context business intelligence and analytics
- Custom fine-tuning for domain-specific performance
Strengths & Limitations
| Strengths | Limitations |
|---|---|
| High-capacity MoE design for strong reasoning and generalization | Higher compute and memory requirements compared to smaller gpt-oss models |
| Optimized activation load for high throughput (12B active parameters) | Latency may increase on single-GPU deployments |
| State-of-the-art performance under native FP4 and FP8 quantization | Fine-tuning recommended for highly specialized enterprise domains |
| Scales across multi-GPU clusters and distributed inference setups | |
| Up to 256K context window with efficient sparse attention | |
| Superior agentic and planning abilities for sequential decision tasks | |
| Built-in support for structured schema-based function calling | |
| Apache 2.0 license enabling commercial and derivative use |
Why Qubrid AI?
- No infrastructure setup — serverless API, pay only for what you use
- OpenAI-compatible — drop-in replacement using the same SDK
- Enterprise-ready — API logs, usage tracking, and team management built in
- Multi-language support — Python, JavaScript, Go, cURL out of the box
- Fast onboarding — get your first response in under 2 minutes
Resources
| Resource | Link |
|---|---|
| 📖 Qubrid Docs | docs.platform.qubrid.com |
| 🎮 Playground | Try in Playground |
| 🔑 API Keys | Get API Key |
| 🤗 Hugging Face | openai/gpt-oss-120b |
| 💬 Discord | Join Community |
Built with ❤️ by Qubrid AI
Frontier models. Serverless infrastructure. Zero friction.
Frontier models. Serverless infrastructure. Zero friction.