SSyncDev

Aion 1.0 Instruct 2.3B

SyncDev model directory · Profile updated Nov 19, 2026

Compare with another model

Aion 1.0 Instruct 2.3B is Microsoft's MIT-licensed on-device small language model and Phi Silica successor for Windows, with a 32K context.

Key specifications

Specification score
25.1
Context window
32.77K
License
MIT

The specification score summarizes recorded features and limits. It is not a benchmark result.

Pricing conditions: Open weights under MIT; no hosted API price.

Inputs and outputs

Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.

Accepts

  • Text input

Produces

  • Text output

Capabilities

Function callingOpen weights

What is Aion 1.0 Instruct 2.3B?

Phi Silica has a successor. Microsoft's Windows Applied Sciences Group published Aion 1.0 Instruct 2.3B on Hugging Face on 18 November 2026, describing it as the foundational small language model for the Windows ecosystem, and said that a further-optimised build, paired with a local content moderation model, will ship as the Windows OS model by the end of January 2027.

So this is a preview of something you will eventually call through Windows itself. The weights, though, are already open under MIT.

What is in the box

Detail Value
Repository microsoft/Aion-1.0-Instruct-2.3B
Parameters ~2.29B, tied input/output embeddings
Layers / hidden size 28 / 2048
Attention GQA, 24 query and 8 KV heads; 21 sliding-window layers (2,048 window) plus 7 full-attention layers
Context 32,768 tokens
Vocabulary 200,029
Precision bfloat16
Knowledge cutoff December 2025
Languages English, German, Spanish, Portuguese, Italian, French, Chinese, Japanese
License MIT

An int4 ONNX build for WebGPU appeared the same day, and Microsoft says further quantised versions will follow the unquantised release.

Running it today

Loading needs trust_remote_code=True, because the architecture ships as custom modelling code, and the checkpoint was exported with transformers 5.x. Turns end with <|end|>, which doubles as the EOS token. Tool calling works by injecting schemas between <|tool|> tags; the model answers with JSON wrapped in <|tool_call|>, and results go back in a user turn prefixed <|tool_response|>.

For the Windows path there is a separate developer validation package. It targets Copilot+ PCs only: Snapdragon ARM64 through QNN, Intel Lunar Lake through OpenVINO, and AMD Ryzen AI through VitisAI. No NPU provider, no fallback; it will not drop to the CPU. That package is for validation, not redistribution, and its output carries invisible watermarking that can surface when pasted into some applications.

Where a 2.3B model earns its keep

Microsoft frames the model around document comprehension, summarisation, rewriting, question answering and chat on the device, with data never leaving the machine. Thirty-two thousand tokens is enough for a long email thread or a multi-page spec. It is not enough for a codebase.

Its publisher-reported scores set expectations honestly: 55.15 on MMLU, 33.35 on MMLU-Pro, 28.18 on GPQA, 74.91 on GSM8K and 83.90 on IFEval. Those numbers say "follows instructions well, knows a moderate amount, reasons shallowly", and Microsoft says as much, warning that it was not trained to excel at world knowledge and may miss detail on complex multi-step reasoning. Multilingual material made up roughly 6% of training data, so expect weaker output outside English.

If you are comparing it with an earlier Microsoft small model such as Phi-4-mini, note that this is a different architecture rather than a refresh. Existing Phi Silica LoRA adapters will not load. New ones have to be trained with Foundry Toolkit.

A quick fit check

  1. Collect fifty prompts your app already sends to a cloud model for rewriting or summarisation.
  2. Run them locally and score the outputs blind against the cloud answers.
  3. Count factual slips separately from style differences. Small models fail on the former.
  4. If you depend on tool calls, check that the JSON parses on every run; do not assume deterministic output, which Microsoft explicitly warns against.

Pass on most rewriting and summary prompts but fail on facts? Then it suits on-device drafting, with a larger model kept for anything that must be correct.

More from the lab is collected on the Microsoft provider page.

How to evaluate Aion 1.0 Instruct 2.3B

Compare the model on the work you need to complete.

Tasks and inputs

Aion 1.0 Instruct 2.3B is listed for Question answering, Summarization, Text generation. It accepts text and produces text. Use the publisher’s input format and preprocessing for your chosen deployment.

Capacity and cost

Documented limits: 32.77K context. Measure a completed task using the prompts, documents and tools your application will use. Include reasoning and retries in the result.

Choosing a deployment

The weights are downloadable under MIT. Check runtime support and size the complete model and working context for your expected concurrency.

Frequently asked questions

The weights are on Hugging Face at `microsoft/Aion-1.0-Instruct-2.3B` under the MIT license, with an int4 ONNX WebGPU build alongside.

These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.

Google

Gemini 4 Neon

Gemini 4 Neon is Google's mid-tier Gemini 4 model for complex coding and analysis, sitting between Hydrogen and Carbon at $3/$15 per 1M tokens.

1.05M context

OpenAI

GPT-6.2 Astra

GPT-6.2 Astra is OpenAI's November 2026 flagship for hard professional, scientific and agentic work, at $8/$40 per million tokens with a 1.05M context.

1.05M context

OpenAI

GPT-6.2 Sol

GPT-6.2 Sol is OpenAI's everyday model for coding, agents and writing: $2/$10 per million tokens, 1.05M context, released 17 November 2026.

1.05M context

Tencent

Hy4 Flash

Tencent Hy4 Flash is the cost-focused companion to Hy4, keeping vision and agent skills at a far lower token price. Availability is still unsettled.

1.05M context

Alibaba

Qwen4 Max

Qwen4 Max (qwen4-max) is Alibaba's first Qwen4-generation flagship: 1M-token context, text, image and video input, thinking on by default, $1.50/$4.50 per 1M.

1M context

Sarvam

Sarvam-105B-1711

Sarvam-105B-1711 is a refreshed open post-train of Sarvam's 105B MoE, tuned for tool calling, instruction following, agents and Indian languages.

— context