GPT-OSS 120B (StackIT)
The transparent alternative: OpenAI-tier intelligence in a community-governed model.

About the Model
GPT-OSS 120B (StackIT) is OpenAI’s 2026 contribution to the open-source community. It provides GPT-4 tier intelligence under an Apache 2.0 license, allowing organizations to run frontier-class AI on their own hardware. It is specifically designed as a "Reasoning Model," natively supporting chain-of-thought and adjustable reasoning effort levels.
Model Key Capabilities
Transparent Reasoning:
Provides full visibility into its chain-of-thought, making it ideal for legal and medical research where "black box" AI is unacceptable.
JSON Mode Precision:
Native support for strict schema enforcement, ensuring it never breaks API integrations.
Technical Documentation:
Specifically fine-tuned on millions of pages of engineering specs to assist with system architecture and R&D.
Quantization-Friendly:
Features native FP4 support, allowing for highly efficient inference on existing enterprise hardware.
Applications & Use Cases
Clinical Summarization:
Processing entire patient histories locally to ensure HIPAA compliance.
Legal Research:
Analyzing sensitive litigation documents without external cloud exposure.
Local Coding Assistants:
Running a high-IQ coding model entirely on a private workstation.
Recomended Models based on your needs

Qwen (DeepMask)
Versatile model with reasoning and tool use. Strong at document and image analysis & multilingual chat.

Qwen3 (StackIT)
Versatile model with reasoning and tool use. Strong at document and image analysis and multilingual chat.

Kimi K2 (DeepMask)
Best for deep reasoning and tool use. Ideal for long, multi-step tasks and document analysis.
Model Specifications
General | |
|---|---|
Model Provider | OpenAI |
Main Use Cases |
|
Intelligence | |
Reasoning Effort | High |
GPQA Diamond | 80.9% |
Memory | |
Max Context | 131K Tokens |
Context Caching | Supported |
Speed | |
Latency (TTFT) | 0.27s |
Throughput | 262 Tokens/sec |
Cost | |
1M Tokens (I/O) | $0.09 / $0.36 |

