MICROSOFT WINDOWS APPLIED SCIENCES GROUP · FIELD VERIFIED
Aion 1.0 Instruct 2.3B
Primary sources last checked on 19 Nov 2026. Verification freshness is tracked separately from factual change, and unknown values remain explicit rather than inferred.
NotableOfficial publisher original checkpoint is publicly accessible at an immutable 40-character revision, with concrete visible weights and a source-specified license. Independent OMW runtime or benchmark performance has not been established; Notable only.Curation method →
Open Model Weights records Aion 1.0 Instruct 2.3B as a approximately 2.29B publisher-reported parameters-parameter text generation / on-device instruction following open-weight model in the Aion family, with Microsoft Windows Applied Sciences Group listed as the developer. The checked source repository provides 1 recognized weight artifacts, format evidence for Safetensors and a verified context window of 32,768 tokens, publisher-declared; not independently tested. Values that are not evidenced in the checked sources remain explicitly undisclosed or unknown.
The repository declares MIT. Commercial use allowed under mit license conditions. Open Model Weights reports the checked license evidence and classification; this is not legal advice.
No base model is declared in the checked standard lineage metadata, so Open Model Weights does not infer one. Training recipe status: publisher-described-not-reproduced; data disclosure status: Publisher describes public and synthetic training sources and roughly six percent multilingual data.
Microsoft Aion 1.0 Instruct is an approximately 2.29B-parameter MIT-licensed local language model supporting eight languages and an advertised 32K context. Its BF16 weights are downloadable now, while an OS-integrated Windows version is a future roadmap item for January 2027. The model profile links an exact original-repository revision and visible files; hardware and benchmark performance remain untested by Open Model Weights.
Official publisher sourcemicrosoft/Aion-1.0-Instruct-2.3BSource ↗
Original weight evidence1 explicitly recorded source-visible Safetensors path(s)Source ↗
License provenancemit — original LICENSE fileSource ↗
Parameter and context claimapproximately 2.29B publisher-reported parameters; 32,768 tokens, publisher-declared; not independently testedSource ↗
Benchmark statusPublisher or external leaderboard claims only; no independent OMW inference or benchmarkSource ↗
WHY IT MATTERS
The pinned official Microsoft repository exposes model.safetensors, custom modeling code, a model card and a standard MIT LICENSE file.
Microsoft specifies 28 layers with hybrid sliding-window and full attention plus grouped-query attention, for lightweight text tasks such as summarization and rewriting.
The model-card quickstart uses trust_remote_code=True; running a downloaded checkpoint is not proof that the planned Windows OS integration is already released.
MODEL-SELECTION VIEW
Strong fit when…
Local document and email summarization with privacy-oriented architecture.
Evaluating eight-language small-model behavior on specific hardware.
Source-pinned research into text assistants without remote hosting.
Important considerations
Small models can hallucinate and struggle on multi-step reasoning, as Microsoft explicitly notes.
Custom source code needs trusted revision pins and code review before execution.
32K context and Windows integration are source claims and roadmap details, not runtime measurements or an existing OS-wide release.
Model purpose and architecture
Microsoft Aion 1.0 Instruct is an approximately 2.29B-parameter MIT-licensed local language model supporting eight languages and an advertised 32K context. Its BF16 weights are downloadable now, while an OS-integrated Windows version is a future roadmap item for January 2027.
The pinned official Microsoft repository exposes model.safetensors, custom modeling code, a model card and a standard MIT LICENSE file.
Pinned public weight artifacts
Microsoft specifies 28 layers with hybrid sliding-window and full attention plus grouped-query attention, for lightweight text tasks such as summarization and rewriting.
The checked immutable source revision is 28902cbfd9f4e0347d6c853aa1301f5d559b2fc0. All recorded primary weight paths are from the official pinned repository tree. A file name is not a measured runtime result.
License and provenance
The model-card quickstart uses trust_remote_code=True; running a downloaded checkpoint is not proof that the planned Windows OS integration is already released.
License evidence: a standard original LICENSE file under mit. The checkpoint license does not automatically cover third-party datasets, application code or generated outputs.
Performance evidence and limitations
These are publisher or independently hosted third-party benchmark claims, not tests Open Model Weights has executed. A fair comparison requires an exact checkpoint, matched prompts, harness, hardware and repeatable samples.
Small models can hallucinate and struggle on multi-step reasoning, as Microsoft explicitly notes.
Deployment and practical evaluation
Appropriate evaluations include: Local document and email summarization with privacy-oriented architecture. Evaluating eight-language small-model behavior on specific hardware. Source-pinned research into text assistants without remote hosting. Source code and quantizations from separate repositories must be checked independently.
Custom source code needs trusted revision pins and code review before execution. 32K context and Windows integration are source claims and roadmap details, not runtime measurements or an existing OS-wide release.
IMPLEMENTATION & EVALUATION
How to evaluate Aion 1.0 Instruct 2.3B before deployment
Primary-source pinned checkpoint evaluation steps; no independent OMW model inference performed.
STEP 01
Pin official checkpoint and artifact paths
Publisher repositories can change after publication.
REPRODUCIBLE ACCEPTANCE TEST
Record the source 40-character revision and verified visible filenames, along with processor/config revision.
WHAT CAN GO WRONG
Mutable tags, missing shards and silent upstream changes.
STEP 02
Review license and code dependencies
Open model weights do not imply unrestricted use of code or all data.
REPRODUCIBLE ACCEPTANCE TEST
Inspect the exact publisher license source, custom modeling code, dependencies and intended use policy.
WHAT CAN GO WRONG
Misclassified commercial license or remote-code execution.
STEP 03
Measure hardware, context and workload
Advertised parameters, context and token rates are not observed system requirements.
REPRODUCIBLE ACCEPTANCE TEST
Publish repeatable measured memory, batch, hardware, latency and context settings.
WHAT CAN GO WRONG
Out-of-memory, unsupported attention operators and overclaimed throughput.
STEP 04
Separate external claims from OMW results
Publisher comparisons have different harnesses.
REPRODUCIBLE ACCEPTANCE TEST
Use held-out tasks and matched baselines before independent ranking.
WHAT CAN GO WRONG
False benchmarking attribution or ungrounded Leading status.
Questions people ask about Aion 1.0 Instruct 2.3B.
Are the original Aion 1.0 Instruct 2.3B model weights publicly available?
Yes. The official source repository microsoft/Aion-1.0-Instruct-2.3B exposes actual Safetensors checkpoint artifacts at the pinned revision 28902cbfd9f4e0347d6c853aa1301f5d559b2fc0.
Is Aion 1.0 Instruct 2.3B independently benchmarked by Open Model Weights?
No. Published scores and throughput descriptions are attributed to their source; OMW has not run model inference or a controlled benchmark.
What license applies to this exact checkpoint?
mit. Evidence is the repository's original license file; follow exact terms and separately inspect dependencies.
Is the model suitable for local deployment?
Original weights are downloadable, but actual fit, operators and performance need hardware-specific testing. Custom source code needs trusted revision pins and code review before execution.
Does Notable mean the model leads an independent benchmark?
No. Notable records a distinct useful, source-verifiable original checkpoint. Leading requires separate matched independent evidence.
03 · PRACTICAL TECHNICAL PROFILE
Deployment, licensing and reproducible evidence.
Inspect weight-memory estimates, model lineage and runtime evidence separately from promotional or untested performance claims.
DEPLOYMENT PROFILE
Weight memory, without pretending it is a deployment guarantee.
These bars visualize weight-only decimal-GB estimates. They exclude KV cache, activations, runtime overhead, optimizer state and sharding.
WEIGHT-ONLY MEMORY
BF16 / FP16~4.6 GB
FP8 / INT8~2.3 GB
INT4~1.1 GB
FIELD EVIDENCE MAP
WeightsVerified
LicenseVerified
Commercial useAllowed
ContextNot declared
FormatsVerified
LineageNot declared
Training recipePublisher described not reproduced
Training dataPublisher describes public and synthetic training sources and roughly six percent multilingual data
RuntimeSource-derived · not runtime-tested
LICENSE & USE
MIT
Commercial use allowed under mit license conditions
Declared licenseMIT
ClassificationAllowed
VerificationLicense file checked
This section summarizes the checked repository evidence. It is not legal advice. Inspect license evidence ↗
LINEAGE & PROVENANCE
What the checked sources disclose — and what they do not.
Open Model Weights does not infer a hidden base model or training corpus when the checked sources do not declare one.
LINEAGE
BASE / PARENTNot declared in model-card metadataUnknown
not inferred
CURRENT RECORDAion 1.0 Instruct 2.3BAion
TRAINING DISCLOSURE
Training recipePublisher described not reproduced
Training dataPublisher describes public and synthetic training sources and roughly six percent multilingual data
FamilyAion
RUNTIME SIGNALS
Compatibility claims stay attached to their evidence level.
Declared and model-card-mentioned runtimes are discovery signals. They are not treated as Open Model Weights runtime tests unless explicitly measured.
MentionedTransformers
MentionedONNX
MentionedGGUF
OMW runtime testNot performed
EVIDENCE TRAIL
Sources and observed history.
This is where the article remains auditable: source links on one side, the registry's observed change history on the other.
Aion 1.0 Instruct 2.3B: technical evidence at a glance
Three evidence-linked visual summaries: parameter scale, illustrative precision memory and verifiable provenance. No performance ranking is implied.
01 / PARAMETER SCALE
Total versus active parameters
Total parameters2.29B
Active parameter count not established.
Repository-derived or publisher-described parameter count
Published total parameters are not a model-quality or performance score.
02 / MEMORY ESTIMATE
Weight-only precision footprint
BF16 / FP164.58 GB
FP8 / INT82.29 GB
INT41.15 GB
Hypothetical weight-only decimal GB, not measured peak VRAM or guaranteed support for each format. KV cache, activations, runtime and sharding are excluded.
Repository SHA 28902cbfd9f4…; a source check is not an inference benchmark or legal clearance. See the per-model evidence JSON for the exact source record.
A verification check does not pretend the model changed. Open Model Weights updates the verification date after a real primary-source check, while factual-update dates move only when retained source evidence changes.
Declared, mentioned and externally evidenced runtimes
Declared librarytransformers
Declared/taggedNone
Mentioned in model cardTransformers, ONNX, GGUF
Artifact signalsNone
External primary runtime evidenceNone
Tested by Open Model WeightsNo — source-derived support only
Open Model Weights separates model-repository evidence from external runtime-release evidence. An external support statement does not change the official model repository's weight formats and does not imply an Open Model Weights inference benchmark.
training recipesource-described-not-independently-reproduced
data disclosurePublisher describes public and synthetic training sources and roughly six percent multilingual data
runtime supportsource-derived-not-runtime-tested
MODELS FOR AI AGENTS · ROLE EVIDENCE
Where this model fits in an agent system
Specialized model. A supporting component. This checkpoint does not, by category alone, serve as the planning and tool-execution controller.
The curated notable tier is not an agent-readiness rating. Model-specific tool-call success, long-horizon task accuracy and compatible runtime configurations are not independently established here.
Source-linked license: MIT (allowed commercial-use classification; see original terms).
Context: 32768 configured tokens.
BF16 weights: 4.58 estimated decimal GB, not runtime VRAM.