Specification

Release date: 11 Aug 2026
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Status: announced
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Openness: proprietary
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Architecture: SigLIP2-so400M vision encoder and Phi-4-mini-instruct (3.8B) language model connected through a lightweight adapter, with task-specific auxiliary heads for classification and visual grounding
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Parameters: 3.8B
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Modalities: image, text
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Input modalities: image, text
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted
Output modalities: text
Source:Microsoft Research — publications & blogT2observed 4 h agomediumLLM-extracted

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

15

Source tiers

T215

Freshest observation

4 h ago

Conflicts

None

Modalities

Modalities: imagetext
Input: imagetext
Output: text

Capabilities

Tool calling
Yes
Microsoft Research — publications & blog · T2
Structured output
Yes
Microsoft Research — publications & blog · T2
Reasoning
Yes
Microsoft Research — publications & blog · T2
Vision
Yes
Microsoft Research — publications & blog · T2
Audio
No
Microsoft Research — publications & blog · T2
Fine-tuning available
Unavailable

No structured attributes yet.

No benchmark results recorded

Results appear when a tier 1–3 source publishes them; we never copy scores without a source.

Current prices

No current prices recorded

Prices appear when a provider publishes a public pricing page we crawl.

Price history

Estimated

Memory need = bytes per parameter (4-bit ≈ 0.5 × 1.15 overhead, 8-bit 1.0, fp16 2.0) + a KV-cache allowance. Not a measurement.

Estimated hardware fit
Hardware	Quantization	Memory	Est. need	Fits
NVIDIA DGX B200	4bit	1,440 GB	2.7 GB	Yes
Apple M3 Ultra	4bit	—	2.7 GB	Yes
Mac Studio (Apple M5 Ultra)	4bit	—	2.7 GB	Yes
AMD Instinct MI325X	4bit	256 GB	2.7 GB	Yes
AMD Instinct MI300X	4bit	192 GB	2.7 GB	Yes
Apple M2 Ultra	4bit	—	2.7 GB	Yes
NVIDIA B200	4bit	180 GB	2.7 GB	Yes
NVIDIA H200	4bit	141 GB	2.7 GB	Yes
NVIDIA H200 NVL	4bit	141 GB	2.7 GB	Yes
Apple M1 Ultra	4bit	—	2.7 GB	Yes
Apple M3 Max	4bit	—	2.7 GB	Yes
Apple M4 Max	4bit	—	2.7 GB	Yes
Mac Studio (Apple M5 Max)	4bit	—	2.7 GB	Yes
MacBook Pro (Apple M5 Max)	4bit	—	2.7 GB	Yes
NVIDIA DGX Spark	4bit	128 GB	2.7 GB	Yes
Apple M2 Max	4bit	—	2.7 GB	Yes
NVIDIA H100 NVL	4bit	94 GB	2.7 GB	Yes
NVIDIA A100 80GB	4bit	80 GB	2.7 GB	Yes
NVIDIA H100 SXM	4bit	80 GB	2.7 GB	Yes
Apple M1 Max	4bit	—	2.7 GB	Yes
Apple M4 Pro	4bit	—	2.7 GB	Yes
Mac mini (Apple M5 Pro)	4bit	—	2.7 GB	Yes
MacBook Pro (Apple M5 Pro)	4bit	—	2.7 GB	Yes
Apple M3 Pro	4bit	—	2.7 GB	Yes
Apple M1 Pro	4bit	—	2.7 GB	Yes
Apple M2 Pro	4bit	—	2.7 GB	Yes
Apple M4	4bit	—	2.7 GB	Yes
iMac (Apple M4)	4bit	—	2.7 GB	Yes
Mac mini (Apple M6)	4bit	—	2.7 GB	Yes
MacBook Air (Apple M5)	4bit	—	2.7 GB	Yes
MacBook Pro (Apple M5)	4bit	—	2.7 GB	Yes
NVIDIA GeForce RTX 5090	4bit	32 GB	2.7 GB	Yes
Apple M2	4bit	—	2.7 GB	Yes
Apple M3	4bit	—	2.7 GB	Yes
NVIDIA GeForce RTX 3090	4bit	24 GB	2.7 GB	Yes
NVIDIA GeForce RTX 4090	4bit	24 GB	2.7 GB	Yes
Apple M1	4bit	—	2.7 GB	Yes

Ancestors 1

Phi-4-mini-instruct (3.8B)

This model

CARE-X

3.8B params

Quantizations 0

None recorded.

Descendants 0

None recorded.

Papers 0

No papers linked yet.

Repositories 0

No repositories linked yet.

As of

Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.

Claim history

15 claims · 15 properties

Release daterelease_date1

Claim history for Release date
Value	Valid from → to	Status	Source	Confidence	Extractor
11 Aug 2026	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Statusstatus1

Claim history for Status
Value	Valid from → to	Status	Source	Confidence	Extractor
announced	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Opennessopenness1

Claim history for Openness
Value	Valid from → to	Status	Source	Confidence	Extractor
proprietary	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Architecturearchitecture1

Claim history for Architecture
Value	Valid from → to	Status	Source	Confidence	Extractor
SigLIP2-so400M vision encoder and Phi-4-mini-instruct (3.8B) language model connected through a lightweight adapter, with task-specific auxiliary heads for classification and visual grounding	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Parametersparameter_count1

Claim history for Parameters
Value	Valid from → to	Status	Source	Confidence	Extractor
3.8B	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Modalitiesmodalities1

Claim history for Modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
image, text	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Input modalitiesmodalities_input1

Claim history for Input modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
image, text	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Output modalitiesmodalities_output1

Claim history for Output modalities
Value	Valid from → to	Status	Source	Confidence	Extractor
text	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Audioaudio1

Claim history for Audio
Value	Valid from → to	Status	Source	Confidence	Extractor
No	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Reasoningreasoning1

Claim history for Reasoning
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Safety notessafety_notes1

Claim history for Safety notes
Value	Valid from → to	Status	Source	Confidence	Extractor
CARE-X is a research model and not a Microsoft product offering or medical device. It has not been cleared or approved by any regulatory authority and is not intended for clinical diagnosis, screening, or patient care. The results described below are retrospective research findings and do not establish the safety, effectiveness, or suitability of CARE-X for any clinical use.	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Structured outputstructured_output1

Claim history for Structured output
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Tool callingtool_calling1

Claim history for Tool calling
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Training data notestraining_data_notes1

Claim history for Training data notes
Value	Valid from → to	Status	Source	Confidence	Extractor
Trained on real-world Indian clinical data from Narayana Health, including rare ICU pathologies and CT-confirmed enlargement conditions. Used de-identified, retrospective clinical data under applicable institutional ethics review.	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Visionvision1

Claim history for Vision
Value	Valid from → to	Status	Source	Confidence	Extractor
Yes	11 Sept 2026 21:08 UTC → current	current	Microsoft Research — publications & blogT2	medium	llm

Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →

11 Aug 2026
New modelModelCARE-XMicrosoft
New model: CARE-X (Microsoft)
microsoft_research
microsoft.com

Full timeline →

Source documents
Source	Document	Type	Tier	Last observed	Snapshots
Microsoft Research — publications & blog	microsoft.com/en-us/research/blog/introducing-care-x-towards-clinically-useful-radiology-vlms-with-auxiliary-supervision-reward-aligned-learning-and-tool-augmented-measurement	news	T1· Official	4 h ago	1

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.