M

Modal

Infrastructure

Serverless GPU platform: a Python decorator puts a function on a GPU, scales it from zero to thousands of containers and bills by the second, aimed at inference, fine-tuning and batch work rather than long-lived clusters.

Total funding$466M

Valuation—

Employees—

Founded—

HeadquartersNew York, USA

StatusPrivate

Products

Selected offerings — the AI products recorded here, not the vendor's full catalogue.

• Modal Infrastructure service Sold on its own

  Serverless GPU compute: a Python decorator puts a function on an accelerator, scales it from zero to thousands of containers and stops billing when it stops running — aimed at inference, fine-tuning and batch jobs rather than reserved clusters.

Inside Modal

• Modal Inference Infrastructure service Included with the parent

  Serve, scale and optimise model inference on Modal's runtime, with sub-second cold starts and autoscaling across regions.

• Modal Training Infrastructure service Included with the parent

  Managed training runs on Modal's fleet, configured in Python alongside the rest of a team's code.

• Modal Sandboxes Infrastructure service Included with the parent

  Isolated, instantly-started containers for running untrusted or agent-generated code at scale — the primitive behind AI app-generation products.

• Modal Notebooks Developer tool Included with the parent

  Hosted notebooks backed by Modal's GPUs, for profiling and experimenting without provisioning a machine.

• Modal Batch Infrastructure service Included with the parent

  Batch execution of large jobs across Modal's fleet, described as one line of code on the product page.

Closest alternatives 4

RRunPodPrivateNamed alongside Modal and Beam as one of the serverless-GPU platforms organizations directly trade off against each other.5 of 5 capabilitiesShared product typeSourced competitorBuy data-centre AI computeCompareBBasetenPrivate · $13BSacra, on record: 'Modal Labs is Baseten's closest competitor' — both are model-serving/inference-hosting platforms.3 of 5 capabilitiesShared product typeSourced competitorServe a model in productionCompareTATogether AIPrivate · $8.3BNamed as a secondary competitor for open-source model hosting/serving, the same job Modal's platform performs.4 of 5 capabilitiesShared product typeSourced competitorServe a model in productionCompareFAFireworks AIPrivate · $17.5BNamed as a secondary competitor for fast inference of open-weight models, the same job Modal's platform performs.3 of 5 capabilitiesDifferent layerSourced competitorServe a model in productionCompare

Similar products 76

How this list is ranked ↗

NGNebius GroupPublic · $61.6BModal and Serverless AI do comparable work on GPU cloud and model hosting; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareRReplicatePrivateModal Inference and Replicate do comparable work on model hosting; both also serve buyers who need to serve a model in production; Replicate's scale not recorded.2 of 5 capabilitiesShared product typeCompareCCrusoePrivate · $10BModal Inference and Crusoe Managed Inference do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (private).4 of 5 capabilitiesShared product typeCompareLLambdaPrivate · $1.5B raisedModal and Lambda Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (private).4 of 5 capabilitiesShared product typeCompareOOllamaPrivate · $88M raisedModal Inference and Ollama do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (private).2 of 5 capabilitiesShared product typeCompareFFalPrivateModal and fal Serverless do comparable work on GPU cloud and model hosting; both also serve buyers who need to buy data-centre AI compute and serve a model in production; Fal's scale not recorded.3 of 5 capabilitiesShared product typeCompareMModularAcquired ($3.1B deal)Modal Inference and Modular Cloud do comparable work on model hosting; both also serve buyers who need to serve a model in production; Modular is acquired, with no independent scale.2 of 5 capabilitiesShared product typeCompareCMComet MLPrivate · $70M raisedModal Notebooks and Comet Artifacts do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (private).1 of 5 capabilitiesShared product typeCompareVVMwareAcquiredModal Inference and VMware Cloud Foundation Private AI Services do comparable work on model hosting; both also serve buyers who need to serve a model in production; VMware is acquired, with no independent scale.2 of 5 capabilitiesShared product typeCompareCCCactus ComputePrivateModal Inference and Cactus both serve buyers who need to serve a model in production; Cactus Compute's scale not recorded.2 of 5 capabilitiesShared product typeCompareW&Weights & BiasesAcquiredModal Inference and W&B Serverless Inference do comparable work on model hosting; both also serve buyers who need to serve a model in production; Weights & Biases is acquired, with no independent scale.4 of 5 capabilitiesShared product typeCompareHFHugging FacePrivate · $4.5BModal Inference and Inference Endpoints both serve buyers who need to serve a model in production; same layer and the same scale band (growth-stage private).3 of 5 capabilitiesShared product typeSame scale bandCompareDDatabricksPrivate · $190BModal Notebooks and Databricks Model Training do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (private).4 of 5 capabilitiesShared product typeCompareCCoreWeavePublic · $48.5BModal and CoreWeave AI Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).3 of 5 capabilitiesShared product typeCompareKEKakao EnterprisePrivateModal Inference and Kubeflow do comparable work on model hosting; both also serve buyers who need to serve a model in production; Kakao Enterprise's scale not recorded.4 of 5 capabilitiesShared product typeCompareUUnslothPrivateModal Notebooks and Unsloth do comparable work on model training; both also serve buyers who need to train or fine-tune a model; Unsloth's scale not recorded.2 of 5 capabilitiesShared product typeCompareHCHuawei CloudPrivateModal Inference and MaaS do comparable work on model hosting; both also serve buyers who need to serve a model in production; Huawei Cloud's scale not recorded.4 of 5 capabilitiesShared product typeCompareNCNAVER CloudPrivateModal and ML expert Platform do comparable work on GPU cloud and model hosting; both also serve buyers who need to buy data-centre AI compute and serve a model in production; NAVER Cloud's scale not recorded; ships API service, application and 1 more rather than the same layer.4 of 5 capabilitiesDifferent layerCompareTTuringPrivate · $2.2BModal Notebooks and RL environments do comparable work on model training; both also serve buyers who need to train or fine-tune a model; ships agent platform and data service rather than the same layer.2 of 5 capabilitiesDifferent layerSame scale bandCompareTTolokaPrivateModal Notebooks and Toloka Train do comparable work on model training; both also serve buyers who need to train or fine-tune a model; Toloka's scale not recorded; ships AI agent, data service and 1 more rather than the same layer.2 of 5 capabilitiesDifferent layerCompareEEleutherAINonprofitModal Notebooks and GPT-NeoX do comparable work on model training; both also serve buyers who need to train or fine-tune a model; EleutherAI is a nonprofit, with no independent scale.1 of 5 capabilitiesShared product typeCompareCCore42PrivateModal and Core42 Sovereign AI Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute; Core42's scale not recorded.2 of 5 capabilitiesShared product typeCompareEElasticPublic · $10.2BModal Inference and Elastic Inference Service do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public); ships agent platform, application and 2 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareCCivitaiPrivate · $5.1M raisedModal Inference and Civitai API do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (private); ships API service and platform rather than the same layer.2 of 5 capabilitiesDifferent layerCompareLLabelboxPrivate · $189M raisedModal Notebooks and Terra do comparable work on model training; both also serve buyers who need to train or fine-tune a model; same layer and the same scale band (growth-stage private).3 of 5 capabilitiesShared product typeSame scale bandCompareNNeo4jPrivate · $2BModal Notebooks and Neo4j Graph Data Science do comparable work on model training; both also serve buyers who need to train or fine-tune a model; ships agent platform, application and 1 more rather than the same layer.1 of 5 capabilitiesDifferent layerSame scale bandCompareTTencentPublic · $510.1BModal and Tencent Cloud HAI do comparable work on GPU cloud and model hosting; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).5 of 5 capabilitiesShared product typeCompareCCoherePrivate · $7BModal Inference and Model Vault do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (private).3 of 5 capabilitiesShared product typeCompareHHuaweiPrivateModal Inference and MindIE do comparable work on model hosting; both also serve buyers who need to serve a model in production; Huawei's scale not recorded.3 of 5 capabilitiesShared product typeCompareAAlibabaPublic · $273.4BModal and Alibaba Cloud Elastic GPU Service do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareAAppenPublicModal Notebooks and AI Data Platform do comparable work on model training; both also serve buyers who need to train or fine-tune a model; Appen's scale not recorded; ships data service and platform rather than the same layer.1 of 5 capabilitiesDifferent layerCompareSSenseTimePublic · $7.7BModal and Elastic Container Platform do comparable work on GPU cloud and model hosting; both also serve buyers who need to buy data-centre AI compute; similar scale (public).4 of 5 capabilitiesShared product typeCompareAAMDPublic · $785.1BModal and AMD Developer Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareCSCerebras SystemsPublic · $44.4BModal Notebooks and Cerebras Training Cloud do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships hardware, model API and 1 more rather than the same layer.2 of 5 capabilitiesDifferent layerCompareIIntelPublic · $463.7BModal and Intel Tiber AI Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).3 of 5 capabilitiesShared product typeCompareOOraclePublic · $454.4BModal and OCI Compute for GPUs do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareMAMistral AIPrivate · $13.8BModal Notebooks and Mistral Forge do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (private); ships AI agent, API service and 3 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareNNVIDIAPublic · $5.1TModal Inference and NVIDIA Dynamo-Triton (Triton Inference Server) do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareAAlphabet (Google)Public · $4.2TModal Notebooks and Vertex AI Workbench do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public).4 of 5 capabilitiesShared product typeCompareHHumainPrivateModal and HUMAIN COMPUTE do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; Humain's scale not recorded.4 of 5 capabilitiesShared product typeCompareHHeliconeAcquiredModal Inference and Helicone AI Gateway do comparable work on model hosting; Helicone is acquired, with no independent scale; ships API service and platform rather than the same layer.1 of 5 capabilitiesDifferent layerCompareBBaiduPublic · $31.6BModal and Baidu Baige AI Computing Platform do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute and serve a model in production; similar scale (public).3 of 5 capabilitiesShared product typeCompareSSplunkAcquiredModal Notebooks and Splunk AI Toolkit do comparable work on model training; Splunk is acquired, with no independent scale.1 of 5 capabilitiesShared product typeCompareLSLM StudioPrivateModal Inference and LM Studio do comparable work on model hosting; both also serve buyers who need to serve a model in production; LM Studio's scale not recorded; ships AI agent and application rather than the same layer.2 of 5 capabilitiesDifferent layerCompareTMThinking Machines LabPrivate · $12BModal Notebooks and Tinker do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (private); ships API service rather than the same layer.2 of 5 capabilitiesDifferent layerCompareDTDell TechnologiesPublic · $360.7BModal Inference and Dell Enterprise Hub do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public).3 of 5 capabilitiesShared product typeCompareALAI21 LabsPrivate · $1.4BModal Notebooks and AI21 Post-Training do comparable work on model training; both also serve buyers who need to train or fine-tune a model; ships API service, agent platform and 2 more rather than the same layer.2 of 5 capabilitiesDifferent layerSame scale bandCompareVVercelPrivate · $9.3BModal Inference and Vercel AI Gateway do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (private).2 of 5 capabilitiesShared product typeCompareTTenstorrentPrivate · $2.6BModal and Tenstorrent Cloud do comparable work on GPU cloud; both also serve buyers who need to buy data-centre AI compute; same layer and the same scale band (growth-stage private).2 of 5 capabilitiesShared product typeSame scale bandCompare44ParadigmPublic · $2BModal Notebooks and 4Paradigm Sage HyperCycle do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships application, hardware and 1 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareMPMeta PlatformsPublic · $1.5TModal Inference and Meta Model API do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public); ships AI agent, agent platform and 2 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareBFBlack Forest LabsPrivate · $3.3BModal Training and Open Weights Licensing do comparable work on model training; both also serve buyers who need to train or fine-tune a model; same layer and the same scale band (growth-stage private).1 of 5 capabilitiesShared product typeSame scale bandCompareDDecartPrivate · $4BModal Inference and Cogito do comparable work on model hosting; both also serve buyers who need to serve a model in production; same layer and the same scale band (growth-stage private).3 of 5 capabilitiesShared product typeSame scale bandCompareLALiquid AIPrivateModal Notebooks and LEAP do comparable work on model training; both also serve buyers who need to train or fine-tune a model; Liquid AI's scale not recorded; ships application and platform rather than the same layer.2 of 5 capabilitiesDifferent layerCompareQQualcommPublic · $174.9BModal Inference and Qualcomm AI Hub both serve buyers who need to serve a model in production; similar scale (public).2 of 5 capabilitiesShared product typeCompareMMicrosoftPublic · $3.7TModal Notebooks and Azure Machine Learning do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public).4 of 5 capabilitiesShared product typeCompareIIBMPublic · $216.6BOverlaps on model inference, model training and workflow automation; similar scale (public).3 of 5 capabilitiesShared product typeCompareSASnorkel AIPrivate · $1.3BModal Notebooks and Snorkel Flow do comparable work on model training; both also serve buyers who need to train or fine-tune a model; ships agent platform, data service and 1 more rather than the same layer.1 of 5 capabilitiesDifferent layerSame scale bandCompareAAmazonPublic · $2.8TModal Inference and Amazon SageMaker AI do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public).4 of 5 capabilitiesShared product typeCompareAMApplied MaterialsPublic · $380.7BModal Notebooks and SmartFactory AI Predictive Metrology do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships application, hardware and 1 more rather than the same layer.2 of 5 capabilitiesDifferent layerCompareKKuaishouPublic · $18.6BModal Inference and Vanchin do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public).3 of 5 capabilitiesShared product typeCompareGGitLabPublic · $8.3BModal Inference and GitLab Duo Agent Platform Self-Hosted do comparable work on model hosting; both also serve buyers who need to serve a model in production; similar scale (public).2 of 5 capabilitiesShared product typeCompareYYandexPublicModal Inference and Yandex DataSphere do comparable work on model hosting; both also serve buyers who need to serve a model in production; Yandex's scale not recorded.3 of 5 capabilitiesShared product typeCompareSSiemensPublic · $244.7BModal Notebooks and Rapidminer AI Studio do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships AI agent, application and 1 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareUUiPathPublic · $7.2BOverlaps on model inference, model training and workflow automation; similar scale (public).3 of 5 capabilitiesShared product typeCompareAAdobePublic · $115.9BModal Notebooks and Adobe Firefly Custom Models do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships AI agent, API service and 2 more rather than the same layer.3 of 5 capabilitiesDifferent layerCompareDDeepchecksAbsorbedModal Notebooks and Deepchecks ML Testing both serve buyers who need to train or fine-tune a model; Deepchecks is absorbed, with no independent scale.Shared product typeCompareIMInsilico MedicinePublic · $4.3BModal Notebooks and MMAI Gym do comparable work on model training; both also serve buyers who need to train or fine-tune a model; similar scale (public); ships application, hardware and 1 more rather than the same layer.1 of 5 capabilitiesDifferent layerCompareOOpenRouterPrivate · $1.3BModal Inference and OpenRouter both serve buyers who need to serve a model in production; same layer and the same scale band (growth-stage private).1 of 5 capabilitiesShared product typeSame scale bandCompareEEncordPrivateModal Notebooks and Encord Data Collection both serve buyers who need to train or fine-tune a model; Encord's scale not recorded; ships AI agent, agent platform and 2 more rather than the same layer.1 of 5 capabilitiesDifferent layerCompareRRunwayPrivate · $5.3BOverlaps on model training and workflow automation; similar scale (private).2 of 5 capabilitiesShared product typeCompareUUniphorePrivate · $2.5BModal Inference and Uniphore Model Layer both serve buyers who need to serve a model in production; ships application, data service and 1 more rather than the same layer.2 of 5 capabilitiesDifferent layerSame scale bandCompareNRNous ResearchPrivate · $70M raisedModal Inference and Hermes Cloud do comparable work on model hosting; similar scale (private).2 of 5 capabilitiesShared product typeCompareNn8nPrivateOverlaps on workflow automation; n8n's scale not recorded; ships application and platform rather than the same layer.1 of 5 capabilitiesDifferent layerCompareSAScale AIPrivate · $29BModal Notebooks and Scale Data Engine both serve buyers who need to train or fine-tune a model; similar scale (private); ships data service and platform rather than the same layer.Different layerCompareHAHume AIPrivateModal Notebooks and Data Solutions both serve buyers who need to train or fine-tune a model; Hume AI's scale not recorded; ships API service, application and 2 more rather than the same layer.Different layerCompare

Closest alternatives are pairs a vendor page itself compares. Similar products are matched from what each product does; nothing in that group is asserted as a rivalry.

Sources

Total funding

Official documentation · 4 Sep 2026

Its own company page: "We've raised over $466M from some of the best investors and product leaders in the industry." Recorded as 466,000,000 — the company's own floor, and the word "over" means the true total may be higher.

Valuation

Checked — none published

No valuation published by the company.

Employees

Not found · 4 Sep 2026

No headcount stated.

Founded

Not found · 4 Sep 2026

No founding year stated on the company page.

Revenue

Checked — none published

Private; no revenue published.

Open Modal in the directory

Something wrong here? Send a correction — quote this company id: modal.