Primary sources last checked on 13 Nov 2026. Verification freshness is tracked separately from factual change, and unknown values remain explicit rather than inferred.
NotableFirst-party OpenAI open-weight reasoning model with broad ecosystem support and Apache-2.0 weights.Curation method →
Open Model Weights records gpt-oss-120b as a 117B total / 5.1B active-parameter text open-weight model in the gpt-oss family, with OpenAI listed as the developer. The checked source repository provides 25 recognized weight artifacts, format evidence for PyTorch, Safetensors and a verified context window of 131,072 tokens. Values that are not evidenced in the checked sources remain explicitly undisclosed or unknown.
The repository declares apache-2.0. Commercial use allowed under apache-2.0 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: training-described-in-model-card; data disclosure status: not-disclosed-in-standard-metadata-or-obvious-model-card-section.
gpt-oss-120b is open-weight reasoning-oriented mixture-of-experts checkpoint with downloadable Apache-2.0 weights. OMW checked the public source repository for 117B observed published parameter values, 131,072 configured context tokens, and a apache-2.0 license declaration. This model family is relevant for reasoning and tool-driven workflows; actual tool-calling behavior depends on prompt templates, schema validation and inference engine support rather than the existence of a published checkpoint alone. Its Notable badge is a separate editorial discovery signal, not a performance measurement by OMW.
Source license statusapache-2.0 (allowed commercial-use classification from checked source)Source ↗
Runtime evidence classPublisher/source mentions only; no OMW execution testSource ↗
WHY IT MATTERS
The repository records a 131,072-token context setting and roughly 117 billion parameter values in published metadata. Model-card parameter descriptions may distinguish active versus total parameters, but OMW does not infer runtime throughput from those numbers.
It is a strong comparison candidate when a team wants a named, first-party Apache-2.0 open-weight model and a reproducible serving environment. Test exact accuracy and operational costs against smaller models instead of equating model size with task quality.
MODEL-SELECTION VIEW
Strong fit when…
It is a strong comparison candidate when a team wants a named, first-party Apache-2.0 open-weight model and a reproducible serving environment. Test exact accuracy and operational costs against smaller models instead of equating model size with task quality.
Evaluations that require open-weight reasoning-oriented mixture-of-experts checkpoint with downloadable Apache-2.0 weights and an inspectable original weight/source revision.
Important considerations
Runtime mentions such as Ollama and vLLM are documentation signals. They are not evidence that OMW executed this checkpoint or that a chosen quantized file will satisfy a particular context window.
Validate conversation template, tool-output JSON adherence, long-context behavior and precision-specific runtime changes with pinned weights and fixed prompts.
Architecture and task-specific behavior
The repository records a 131,072-token context setting and roughly 117 billion parameter values in published metadata. Model-card parameter descriptions may distinguish active versus total parameters, but OMW does not infer runtime throughput from those numbers.
This model family is relevant for reasoning and tool-driven workflows; actual tool-calling behavior depends on prompt templates, schema validation and inference engine support rather than the existence of a published checkpoint alone.
Verified configuration versus usage claims
The primary registry record identifies gpt-oss-120b as Text. The checked configuration lists 131,072 tokens. This value is a declared configuration bound, not a measured practical prompt limit. This page treats a checked config field, the model-card prose and a deployment benchmark as three distinct kinds of evidence.
In the current checked snapshot, 117B observed published parameter values are recorded with a source repository revision. Those weight-parameter values do not determine how much GPU memory, active compute or KV-cache is required by a real application; sparse activation, quantization and runtime overhead each need separate verification.
What the actual weight repository contains
OMW checked 25 recognized weight artifacts in PyTorch, Safetensors formats. The 117B observed published parameter values come from the published repository data, not from a newly executed OMW performance test. A file count is not a checksum audit, a full weight-binary comparison or a statement about throughput.
The source revision and exact listed weight files can be inspected through the linked OMW evidence API and the official model repository. Before distributing or deploying a model, pin the precise commit and independently validate downloaded artifact checksums where available. An upstream revision change must not automatically be described as changed weight bytes.
License and intended commercial deployment
The source declares apache-2.0; its commercial classification is allowed under the stated license conditions. Read the actual terms and any attribution obligations before shipping. Source observation does not provide legal advice.
Runtime mentions such as Ollama and vLLM are documentation signals. They are not evidence that OMW executed this checkpoint or that a chosen quantized file will satisfy a particular context window. Keep publisher license declarations separate from third-party interpretations, custom usage agreements and any deployment-specific regulatory obligations. Where a permission is unknown, the registry intentionally reports unknown rather than guessing a permissive answer.
Runtime, hardware and reproduction limits
The source record includes transformers, Ollama, Transformers, vLLM as declared libraries or model-card mentions. This is provenance-aware compatibility information only: OMW has not executed a model load, generated outputs or benchmarked the exact checkpoint on a specified GPU. Runtime support for a particular version, precision and processor remains to be tested.
Validate conversation template, tool-output JSON adherence, long-context behavior and precision-specific runtime changes with pinned weights and fixed prompts. The generic weight-only memory estimate in the model snapshot does not include multimodal processors, activation buffers, temporary tensors, cache growth, device placement or kernel availability; these can dominate real deployments.
When to choose this model and how to compare it
It is a strong comparison candidate when a team wants a named, first-party Apache-2.0 open-weight model and a reproducible serving environment. Test exact accuracy and operational costs against smaller models instead of equating model size with task quality.
For a fair comparison, choose another model from the same task family and freeze the task dataset, prompt or preprocessing settings, tested checkpoint revisions and compute budget. Runtime mentions such as Ollama and vLLM are documentation signals. They are not evidence that OMW executed this checkpoint or that a chosen quantized file will satisfy a particular context window. A Leading or Notable label remains a discovery cue rather than a universal model ranking or an OMW benchmark result.
IMPLEMENTATION & EVALUATION
How to evaluate gpt-oss-120b before deployment
Use these as proposed acceptance tests rather than OMW-confirmed performance results. The objective is to establish whether the exact gpt-oss-120b revision is legal to use and deployable for a specific workload.
STEP 01
Freeze model artifacts and license evidence
Record the official gpt-oss repository SHA, recognized PyTorch, Safetensors artifacts and the apache-2.0 license evidence before attempting an installation.
REPRODUCIBLE ACCEPTANCE TEST
An immutable source revision, artifact manifest, checksums where available and dated license record are saved together.
WHAT CAN GO WRONG
Mutable model branches or unreviewed third-party weight conversions can silently change the experiment or applicable terms.
STEP 02
Reproduce the documented task interface
The repository records a 131,072-token context setting and roughly 117 billion parameter values in published metadata. Model-card parameter descriptions may distinguish active versus total parameters, but OMW does not infer runtime throughput from those numbers.
REPRODUCIBLE ACCEPTANCE TEST
The intended Text workload runs under a documented processor or tokenizer path with exact dependency versions, or the failure is recorded.
WHAT CAN GO WRONG
Runtime mentions such as Ollama and vLLM are documentation signals. They are not evidence that OMW executed this checkpoint or that a chosen quantized file will satisfy a particular context window.
STEP 03
Measure deployment and model-specific outputs
Validate conversation template, tool-output JSON adherence, long-context behavior and precision-specific runtime changes with pinned weights and fixed prompts.
REPRODUCIBLE ACCEPTANCE TEST
Raw evaluation inputs, response or media outputs, memory high-water mark, request latency and error logs are retained and the environment can be reconstructed.
WHAT CAN GO WRONG
Publisher benchmark claims and parameter-derived estimates are not interchangeable with measurements on the chosen hardware.
STEP 04
Compare against a relevant verified alternative
It is a strong comparison candidate when a team wants a named, first-party Apache-2.0 open-weight model and a reproducible serving environment. Test exact accuracy and operational costs against smaller models instead of equating model size with task quality.
REPRODUCIBLE ACCEPTANCE TEST
Run the same tasks and quality rubric against Qwen3.8-27B and Olmo-3-7B-Think, with source revisions and test limitations disclosed.
WHAT CAN GO WRONG
A different license, modality, processor or sampling budget makes a naive head-to-head score misleading.
gpt-oss-120b is open-weight reasoning-oriented mixture-of-experts checkpoint with downloadable Apache-2.0 weights. The repository records a 131,072-token context setting and roughly 117 billion parameter values in published metadata. Model-card parameter descriptions may distinguish active versus total parameters, but OMW does not infer runtime throughput from those numbers.
What is the verified context length for gpt-oss-120b?
The checked configuration lists 131,072 tokens. This value is a declared configuration bound, not a measured practical prompt limit.
How big are the published gpt-oss-120b weights?
117B observed published parameter values; 25 recognized repository weight entries (PyTorch, Safetensors). This is a weight-metadata observation, not a GPU memory measurement.
Can gpt-oss-120b be used commercially?
The source declares apache-2.0; its commercial classification is allowed under the stated license conditions. Read the actual terms and any attribution obligations before shipping. Source observation does not provide legal advice.
Which inference runtimes support gpt-oss-120b?
The checked model documentation mentions transformers, Ollama, Transformers, vLLM. OMW has not run this checkpoint under those runtimes; compatibility must be reproduced for the selected revision.
How should gpt-oss-120b be benchmarked?
Validate conversation template, tool-output JSON adherence, long-context behavior and precision-specific runtime changes with pinned weights and fixed prompts. Keep environment, source commit, data and evaluation method fixed, and distinguish publisher results from your own repeatable tests.
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 / FP16~233.7 GB
FP8 / INT8~116.8 GB
INT4~58.4 GB
FIELD EVIDENCE MAP
WeightsVerified
LicenseVerified
Commercial useAllowed
ContextVerified
FormatsVerified
LineageNot declared
Training recipeTraining described in model card
Training dataNot disclosed in standard metadata or obvious model card section
RuntimeSource-derived · not runtime-tested
LICENSE & USE
apache-2.0
Commercial use allowed under apache-2.0 conditions
Declared licenseapache-2.0
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 RECORDgpt-oss-120bgpt-oss
TRAINING DISCLOSURE
Training recipeTraining described in model card
Training dataNot disclosed in standard metadata or obvious model card section
Familygpt-oss
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.
Declaredtransformers
MentionedOllama
MentionedTransformers
MentionedvLLM
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.
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 parameters116.8B
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 / FP16233.7 GB
FP8 / INT8116.8 GB
INT458.41 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 b5c939de8f75…; 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.
Weight-only decimal-GB estimate from total parameter count; excludes KV cache, activations, runtime overhead, optimizer state and sharding.
04 · FORMATS
Formats & precision evidence
PyTorchSafetensors
BF16: not observedFP16: not observedFP8: not observedINT8: not observedINT4: not observedGGUF: not observed
“Not observed” means no matching repository artifact/name/dtype signal was found; it does not prove no third-party conversion exists.
05 · LINEAGE
Base model & family
Familygpt-oss
Variant120B
Declared base modelNot declared in model-card metadata
RelationNot declared
06 · TRAINING ASSETS
Recipe & data disclosure
Training recipetraining-described-in-model-card
Data disclosurenot-disclosed-in-standard-metadata-or-obvious-model-card-section
EvaluationsNot classified by automated verifier
Intermediate checkpointsNot classified by automated verifier
07 · RUNTIME SUPPORT
Declared, mentioned and externally evidenced runtimes
Declared librarytransformers
Declared/taggedtransformers
Mentioned in model cardOllama, Transformers, vLLM
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.
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Field-by-field verification
Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.
Field-by-field verification
Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Repository revision check
Official Hugging Face repository SHA unchanged; prior field evidence retained.
Field-by-field verification
Checked Hugging Face API metadata, exact repository file list, model card and config when accessible.
Registry record created
Previous verification level: verified.
licenseverified
commercial useclassified
contextverified
weight filesverified
formatsverified
base modelnot-declared
training recipetraining-described-in-model-card
data disclosurenot-disclosed-in-standard-metadata-or-obvious-model-card-section
runtime supportsource-derived-not-runtime-tested
MODELS FOR AI AGENTS · ROLE EVIDENCE
Where this model fits in an agent system
Text reasoning model. A candidate for a reasoning or coordination role. No end-to-end agent performance is inferred from its model category.
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: apache-2.0 (allowed commercial-use classification; see original terms).
Context: 131072 configured tokens.
BF16 weights: 233.66 estimated decimal GB, not runtime VRAM.
These records are separately sourced. They do not replace independent model-plus-runtime tests.
Tool calling · publisher-documented — OpenAI documents native function calling, structured outputs and the required Harmony response format for the gpt-oss models. Original source ↗
Ollama · runtime catalog documented — Official Ollama library serves its own MXFP4 build; do not equate its artifact hash or memory requirement with the Hugging Face weight file. Original source ↗