Microsoft

Multinational tech company; primary infrastructure partner and largest investor in OpenAI, operator of Microsoft Research, and ships Copilot products across Windows, Office, and GitHub.

Visit

Type: frontier lab
HQ: Redmond, WA, USA
Founded: 1975
Funding: Public (NASDAQ: MSFT); ~$3T market cap
Website: microsoft.com
GitHub: github.com/microsoft
Twitter: twitter.com/Microsoft

Cite

Notes

Only stored in your browser.

Evals

Tools

Models

Papers

Boards

People

Models

Ranked by Artificial Analysis Intelligence Index

#	Model	Released	Intelligence
1	Phi 4	12 Dec 2024	10.4
2	WizardLM (13B family)	Apr 2023	-
3	Wizardlm 70B	-	-
4	Phi 3 Medium 4k Instruct	-	-
5	Phi 3 Small 8k Instruct	-	-
6	Phi 3 Mini 4k Instruct June 2024	-	-
7	Phi 3 Mini 128k Instruct	-	-
8	Phi 3.5 Vision Instruct	-	-
9	Phi 3 Vision 128k Instruct	-	-
10	Wizardlm 13B	-	-
11	Phi 3 Mini 4k Instruct	-	-

11 / 11 models

Evals

CTI-REALM: Cyber Threat Intelligence Detection Rule Development Benchmark

Evaluates AI systems' ability to analyze cyber threat intelligence and develop comprehensive detection capabilities through a realistic 5-subtask workflow: MITRE technique mapping, data source discovery, Sigma rule generation, KQL

ActiveCybersecurity

ALFWorld

Embodied household-task benchmark that aligns TextWorld text commands with ALFRED 3D scenes, testing whether agents can transfer from abstract text policies to grounded execution.

ActiveEmbodiedPlanningTool CallingAgentic

Tools

WizardLM Evol-Instruct

Microsoft's "Evol-Instruct" recipe - automatically rewriting simple instructions into harder, more diverse ones using an LLM evolver.

SFT DatasetCode GenerationMathInstruction Following