Agnes 2.5 Pro Alpha
Agnes 2.5 Pro Alpha by Sapiens AI is a proprietary reasoning model for coding, scientific reasoning, long-context analysis, and agentic terminal workflows.
Model Overview
Capabilities, design details, and architectural traits
Agnes 2.5 Pro Alpha - proprietary reasoning model
Agnes 2.5 Pro Alpha is a proprietary reasoning model from Sapiens AI, paired with a commercial stable release called Agnes 2.5 Pro. It targets production workloads that need deeper reasoning than flash-class models, with emphasis on coding, science and math, long-context understanding, and agentic terminal tasks.
| Trait | Detail |
|---|---|
| Dual API compatibility | Available through an OpenAI-compatible Chat Completions API, with a thinking field for enabling Thinking mode in Anthropic-compatible requests. |
| Agentic terminal focus | Designed for agentic coding workflows and terminal tasks. |
| Scientific reasoning depth | Intended for multi-step scientific and knowledge-intensive analysis. |
| Multimodal input | Accepts text and image URL inputs for visual analysis and multimodal reasoning; output is text only. |
| Infinite context window | Listed with a context window of ∞+ tokens for long documents and extended conversations. |
Commercial and stable split
Agnes 2.5 Pro is the commercial stable release. Both share the same base integration pattern and model identifier (agnes-2.5-pro).
Benchmark Performance
Independent evaluations · Artificial Analysis
Accuracy & Capability Details
Compare Models Side-by-Side
Evaluate specifications, pricing, and independent benchmark indices
| Model Details | |||
|---|---|---|---|
| General Info | |||
| Provider | Sapiens AI | Anthropic | Anthropic |
| Release Date | July 24, 2026 | September 22, 2026 | September 28, 2026 |
| Knowledge Cutoff | - | - | Jun 2026 |
| Context & Limits | |||
| Context Window | 262K | 1M Best Context Window | 1M Best Context Window |
| Pricing (per 1M tokens) | |||
| Input Pricing | $0.45 Best Input Pricing | $4 | $2 |
| Output Pricing | $0.90 Best Output Pricing | $20 | $10 |
| Modalities | |||
| Inputs | textimage | textimagefile | textimagefile |
| Outputs | text | text | text |
| Benchmarks (0-100) | |||
| Intelligence Index | 26.8 | 57.6 Best Intelligence Index | 56.0 |
| Coding Index | 58.8 | - | - |
| Agentic Index | 29.4 | - | - |
Humanity's Last Exam
Extremely difficult logical reasoning and knowledge.
Long Context Reasoning
Logical reasoning over long context windows.
SciCode Benchmark
Scientific coding and mathematical modeling.
Independent evaluation data provided by Artificial Analysis. To view the latest benchmarks and full details, visit their official site.
Explore more from Sapiens AI
Other models by Sapiens AI
Top AI Models
Leading alternatives by intelligence score
