Microsoft
Microsoft has 47 models in Epoch AI's notable models list, 2 of them released since January 2025; the largest training run with a compute estimate is Megatron-Turing NLG 530B, at 8.6e23 FLOP. Its published safety framework is the Frontier Governance Framework, and its one indexed release since January 2025 had a safety card on the day of release.
- 8.6e23 FLOP largest training run, Megatron-Turing NLG 530B, 2021 Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 28 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/.
- 2 notable models released since January 2025
- 1 of 1 releases since January 2025 with a system card on the day Release Safety Disclosure Index This site Our own reading of the system card, model card or safety report each developer published for each release Epoch AI lists from 2025, joined to Epoch's list. Every cell is what a developer says about its own model under its own framework: whether a card came out with the release, whether it reports dangerous-capability results, which government institutes it names as pre-release testers, and whether it says a risk threshold was reached. None of it is verified. A release Epoch adds after the last review is shown as awaiting review. Cite AI Safety Tracker, "Release Safety Disclosure Index". https://aisafetytracker.org/alignment/#releases. Retrieved 18 October 2026. Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 18 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/.
- 7 sites it owns in Epoch's data-centre hub, and 4 where it is a named user
Models
Every model Epoch AI judges notable from Microsoft, by year, with the largest training run each year where Epoch has a compute estimate.
| Year | Models | Models, relative | Largest run that year |
|---|---|---|---|
| 2016 | 4 | ResNet-200, 3.0e19 FLOP | |
| 2017 | 4 | ISS, 3.4e15 FLOP | |
| 2018 | 2 | - | |
| 2019 | 1 | - | |
| 2020 | 2 | Turing-NLG, 1.6e22 FLOP | |
| 2021 | 11 | Megatron-Turing NLG 530B, 8.6e23 FLOP | |
| 2022 | 6 | BEIT-3, 7.0e19 FLOP | |
| 2023 | 4 | LLaVA 1.5, 7.8e22 FLOP | |
| 2024 | 0 | - | |
| 2025 | 0 | - | |
| 2026 | 2 | - |
Most recent
| Published | Model | Domain | Released as | Training compute |
|---|---|---|---|---|
| 2026-06-02 | MAI-Image-2.5 | Image generation | API only | - |
| 2026-06-02 | MAI-Code-1-Flash | Language | API only | - |
| 2023-11-05 | LLaVA 1.5 | Multimodal | Open weights | 7.8e22 |
| 2023-10-26 | CODEFUSION (Python) | Language | Unreleased | 7.9e18 |
| 2023-04-17 | LLaVA | Multimodal | Open weights | 7.8e22 |
| 2023-01-05 | VALL-E | Audio | Unreleased | 1.0e19 |
| 2022-08-22 | BEIT-3 | Multimodal | Open weights | 7.0e19 |
| 2022-06-13 | MetaLM | Multimodal | Unreleased | - |
| 2022-06-05 | Diffusion-GAN | Image generation | Open weights | - |
| 2022-05-04 | DeBERTaV3large + KEAR | Language | Unreleased | - |
Earlier models 37
| Published | Model | Domain | Released as | Training compute |
|---|---|---|---|---|
| 2022-03-21 | Segatron-XL large, M=384 + HCP | Language | Unreleased | 2.6e19 |
| 2022-03-01 | DeepNet | Language | Unreleased | - |
| 2021-12-03 | T-NLRv5 XXL | Language | Unreleased | - |
| 2021-11-24 | NÜWA | Multimodal | Unreleased | 7.2e21 |
| 2021-11-22 | Florence | Vision | Unreleased | 4.8e22 |
| 2021-11-18 | Swin Transformer V2 (SwinV2-G) | Vision | Open weights | 1.1e21 |
| 2021-10-11 | Megatron-Turing NLG 530B | Language | Unreleased | 8.6e23 |
| 2021-09-28 | Turing ULRv5 | Language | API only | 2.9e22 |
| 2021-09-21 | TrOCR | Vision | Open weights | - |
| 2021-09-04 | MEB | Search | API only | - |
| 2021-06-28 | Adaptive Input Transformer + RD | Language | Unreleased | 8.6e19 |
| 2021-06-10 | DeBERTa | Language | Open weights | 2.6e22 |
| 2021-02-09 | DLWP | Earth science | Unreleased | 5.7e18 |
| 2020-02-13 | Turing-NLG | Language | Unreleased | 1.6e22 |
| 2020-02-07 | Theseus 6/768 | Language | Open weights | - |
| 2019-01-31 | MT-DNN | Language | Open weights | - |
| 2018-09-18 | AWD-LSTM-MoS + dynamic evaluation (WT2, 2018) | Language | Unreleased | - |
| 2018-03-01 | Chinese - English translation | Language | Not recorded | - |
| 2017-09-15 | ISS | Language | Unreleased | 3.4e15 |
| 2017-07-30 | GSM | Language | Not recorded | - |
| 2017-06-13 | HRA | Games | Not recorded | - |
| 2017-05-08 | Mnemonic Reader | Language | Not recorded | - |
| 2016-09-17 | ResNet-1001 | Vision | Not recorded | - |
| 2016-09-17 | ResNet-200 | Vision | Unreleased | 3.0e19 |
| 2016-09-12 | MS-ensemble-speech-recognition | Speech | Unreleased | - |
| 2016-06-21 | R-FCN | Vision | Not recorded | 7.2e17 |
| 2015-12-10 | ResNet-152 (ImageNet) | Vision | Not recorded | 1.0e19 |
| 2015-12-10 | ResNet-110 (CIFAR-10) | Vision | Not recorded | - |
| 2015-12-10 | ResNet-101 (ImageNet) | Vision | Open weights | 7.0e18 |
| 2015-06-04 | Faster R-CNN | Vision | Open weights | - |
| 2015-04-30 | Fast R-CNN | Vision | Not recorded | - |
| 2015-02-06 | MSRA (C, PReLU) | Vision | Not recorded | 2.4e19 |
| 2014-09-18 | Deeply-supervised nets | Vision | Not recorded | - |
| 2014-06-18 | SPPNet | Vision | Not recorded | 3.4e18 |
| 2012-12-01 | RNN+LDA+KN5+cache | Language | Unreleased | - |
| 2012-07-27 | Context-dependent RNN | Language | Not recorded | - |
| 2005-08-07 | RankNet | Search | Unreleased | 3.5e12 |
Notability is Epoch's editorial judgement, and compute figures are its estimates, most of them not disclosed by the developer. Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 28 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/. Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 28 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/.
Safety framework and release disclosures
The frontier safety framework Microsoft has published, and what it published at each release since January 2025.
- Framework
- Frontier Governance Framework since 2025
- Its own scale
- Risk levels per capability: low, medium, high, critical
- Risk domains
- CBRN, cyber, AI R&D, loss of control, manipulation
- External evaluation
- committed to
- Commits to stop
- yes: Commits to not deploying, or not continuing, until mitigations are in place.
Revised in February 2026 to align with the EU Code of Practice and the Californian and New York frontier AI laws, adding loss of control and harmful manipulation. Commits to pause development and deployment of a model whose risk cannot be sufficiently mitigated. Covers models in scope of those laws, and Microsoft's substantial fine-tunes of other developers' models.
- 1 of 1 with a system card on the day
- 0 that no card names
- 0 naming a government institute that tested before release
- 0 stated to have reached a threshold in its own framework
| Released | Model | Card | Results | Government tested | Threshold |
|---|---|---|---|---|---|
| 2026-06-02 | MAI-Code-1-Flash The card reports secure-coding evaluations, which are not dangerous-capability ones. | at release | none | none named | not stated Names no threshold |
Every entry is Microsoft's own statement about its own models under its own framework. The rules behind each column are on the alignment page. Frontier Safety Framework Index This site Our own reading of each developer's published safety framework and model cards. Every field is what the developer says about itself: these are self-assessments under self-written frameworks with no common scale and no external audit, and they are indexed here for exactly that reason. Not a ranking, and not a verification. Cite AI Safety Tracker, "Frontier Safety Framework Index". https://aisafetytracker.org/alignment/#frameworks. Retrieved 18 October 2026. Release Safety Disclosure Index This site Our own reading of the system card, model card or safety report each developer published for each release Epoch AI lists from 2025, joined to Epoch's list. Every cell is what a developer says about its own model under its own framework: whether a card came out with the release, whether it reports dangerous-capability results, which government institutes it names as pre-release testers, and whether it says a risk threshold was reached. None of it is verified. A release Epoch adds after the last review is shown as awaiting review. Cite AI Safety Tracker, "Release Safety Disclosure Index". https://aisafetytracker.org/alignment/#releases. Retrieved 18 October 2026. Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 18 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/.
Its models against the compute thresholds
How many of Microsoft's models with a compute estimate sit at or above each training-compute threshold written into an instrument, and the first to cross it.
| Instrument | Threshold | Status | Its models above | First of them to cross |
|---|---|---|---|---|
| Regulation (EU) 2024/1689, Article 51(2) European Union | 1e25 | in force | 0 | - |
| SB 53, Transparency in Frontier Artificial Intelligence Act United States (California) | 1e26 | in force | 0 | - |
| Enforcement Decree of the AI Basic Act, Article 23 South Korea | 1e26 | in force | 0 | - |
| Executive Order 14110, section 4.2 United States (federal) | 1e26 | revoked | 0 | - |
| SB 1047, Safe and Secure Innovation for Frontier AI Models Act United States (California) | 1e26 | vetoed | 0 | - |
| RAISE Act, Responsible AI Safety and Education Act United States (New York) | 1e26 | enacted | 0 | - |
Arithmetic on Epoch's compute estimates against the number in each instrument, not a legal finding about any model. The full timeline is on the capability page. Compute Threshold Index This site Training-compute thresholds written into law, executive action or published policy, with the text that sets each one. Whether a given model crosses a threshold here is arithmetic on Epoch's compute estimates, most of which are estimates rather than disclosures, and is not a legal finding about any model or developer. Cite AI Safety Tracker, "Compute Threshold Index". https://aisafetytracker.org/capability/#thresholds. Retrieved 23 October 2026. Notable AI Models Epoch AI Inclusion is editorial ("notability criteria"), not exhaustive. Compute figures are frequently estimates; check the per-row confidence and notes columns before treating any value as measured. Cite Epoch AI, "Notable AI Models". https://epoch.ai/data/ai-models. Retrieved 23 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/.
Datacentres
Sites in Epoch AI's data-centre hub that Microsoft owns or is named as using, largest planned IT power first. Those it owns plan 4.1 GW of IT power between them.
| Site | Role | Country | Stage | IT power now | Planned |
|---|---|---|---|---|---|
| Microsoft Fairwater Wisconsin | owner, confident | United States | Expanding | 369 MW | 2.3 GW |
| Crusoe Abilene Expansion | owner, confident | United States | Under construction | - | 672 MW |
| Microsoft Fairwater Atlanta | owner, confident | United States | Operating | 636 MW | 636 MW |
| Start Campus Sines Data Campus | named user, confident | Portugal | Expanding | 33 MW | 233 MW |
| Microsoft Goodyear | owner, confident | United States | Operating | 202 MW | 202 MW |
| Microsoft Project Osmium | owner, confident | United States | Operating | 190 MW | 190 MW |
| Nebius Mantsala | named user, speculative | Finland | Expanding | 75 MW | 145 MW |
| Microsoft Narvik Norway | named user, likely | Norway | Under construction | - | 92 MW |
| Microsoft SAT40 | owner, confident | United States | Operating | 75 MW | 75 MW |
| Microsoft-Nebius New Jersey | named user, likely | United States | Operating | 50 MW | 50 MW |
| Microsoft SAT14 | owner, confident | United States | Operating | 47 MW | 47 MW |
Power is Epoch's estimate from imagery and permits, IT power where it states one. Confidence is Epoch's own label for each owner or user. AI Data Centers Epoch AI The largest AI sites Epoch has found, mostly in the United States. Epoch estimated in September 2026 that they hold 44% of the world's deployed AI compute, with a 90% interval of 23% to 81%. Power, compute and cost are Epoch's estimates from satellite imagery, permits and company statements; four in five IT power estimates fall within a factor of 1.4 of the true value. The owner is the owner of the chips, not necessarily of the building. Capacity after the date of retrieval is Epoch's projection. Coordinates are read from the hub's map page, which publishes them; the downloads do not. Cite Epoch AI, "AI Data Centers". https://epoch.ai/data/ai-data-centers. Retrieved 28 October 2026 via AI Safety Tracker, https://aisafetytracker.org/developers/microsoft/. AI Datacentre Index This site AI datacentre sites that Epoch AI's hub does not hold, recorded from the statements of developers, named tenants and planning authorities. Australia only so far. A site is included only where a primary source says it is built for AI compute or names the AI deployment inside it. Capacity is as the developer states it, IT or facility power as labelled, and is not an estimate; most of it is proposed and not yet approved. Coordinates are this site's placement of each stated address or area. Cite AI Safety Tracker, "AI Datacentre Index". https://aisafetytracker.org/datacentres/#added. Retrieved 28 October 2026.
Energy and emissions
What Microsoft has published about the energy behind its models, and what permits and press reporting add.
Microsoft is not in the Environmental Disclosure Index.
- 100% of electricity matched with renewable purchases over FY2025 , as the company reports it
"Published" means the developer's own statement; "reported by others" means a permit, a regulator or the press. A company's clean-energy claims cover its whole operation, never one model. Environmental Disclosure Index This site Our own reading of what frontier developers and the operators of their data centres have published about training energy, emissions, power sources and energy per prompt, with permits and press reporting marked separately from company statements. Most cells record an absence. Company figures are self-reported and mostly unaudited, and market-based emissions reflect contracts rather than the grid at the hour of use. Cite AI Safety Tracker, "Environmental Disclosure Index". https://aisafetytracker.org/environment/. Retrieved 23 October 2026. Environmental Disclosure Index This site Our own reading of what frontier developers and the operators of their data centres have published about training energy, emissions, power sources and energy per prompt, with permits and press reporting marked separately from company statements. Most cells record an absence. Company figures are self-reported and mostly unaudited, and market-based emissions reflect contracts rather than the grid at the hour of use. Cite AI Safety Tracker, "Environmental Disclosure Index". https://aisafetytracker.org/environment/. Retrieved 23 October 2026.
Usage
The token volumes and user counts Microsoft has stated, each in its own scope.
- 42 trillion tokens a month on average, Microsoft, Foundry, as of June 2025 source
- 900 million monthly users, Microsoft AI features, as of October 2025 source
- 150 million monthly users, Microsoft Copilot family, as of October 2025 source
- 50 million users, GitHub Copilot, as of July 2026 source
Volumes are normalised to a month and are comparable only within one company's own scope. Every series with its scope is on the usage page. AI Usage Disclosure Index This site Our own compilation of the token volumes, user counts and shares of use that companies and studies have put on the record, each linked to where it was stated. Scopes, units and token definitions differ between companies, so no two figures are comparable without the scope printed beside each. Figures known only through press reporting are marked as such. Not a market share estimate, and not a measure of any company's actual volume beyond what it chose to state. Cite AI Safety Tracker, "AI Usage Disclosure Index". https://aisafetytracker.org/usage/. Retrieved 23 October 2026. AI Usage Disclosure Index This site Our own compilation of the token volumes, user counts and shares of use that companies and studies have put on the record, each linked to where it was stated. Scopes, units and token definitions differ between companies, so no two figures are comparable without the scope printed beside each. Figures known only through press reporting are marked as such. Not a market share estimate, and not a measure of any company's actual volume beyond what it chose to state. Cite AI Safety Tracker, "AI Usage Disclosure Index". https://aisafetytracker.org/usage/. Retrieved 23 October 2026.
Not recorded for Microsoft: an entry in the Environmental Disclosure Index.