Mistral AI model product
Shieldstral 1.0
Shieldstral is a 3B open-weight multimodal safety classifier from Mistral.
Updated Aug 17, 2026. Default version: Shieldstral 1.0 (3B)
LLMBoard Score0.0Shieldstral 1.0 (3B)
Coverage20%1 benchmark families
Context windowN/ATokens
Official input priceN/AOfficial price unavailable
Shieldstral 1.0 Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Shieldstral 1.0 Benchmark Results
Benchmark scores for Shieldstral 1.0 (3B).
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkXSTest | Score94.6% | Rank03 | Participants4 | Percentile33.3% | EvidenceC | Evaluated |
Shieldstral 1.0 Arena Results
Preference and agent-evaluation results for the default version.
No Arena results
The default version does not have a matching Arena result yet.
Shieldstral 1.0 Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- N/A
- Tracked offerings
- 0
No provider prices
The default version has no current input or output token prices.
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Shieldstral 1.0 Runtime Performance
Provider-specific output speed and catalog latency for Shieldstral 1.0 (3B). Runtime does not affect the capability score.
No runtime data
Browse runtime rankingsNo provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Shieldstral 1.0 Specifications
Technical details for the model's default version.
- Version
- Shieldstral 1.0 (3B)
- Released
- Aug 4, 2026
- Knowledge cutoff
- Unknown
- Parameters
- 3B
- Context window
- N/A
- Max output
- N/A
- Inputs
- image, text
- Outputs
- text
- Open weights
- No
- License
- Apache 2.0
Shieldstral 1.0 Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionShieldstral 1.0 (3B) | Released | LLMBoard0.0 | Parameters3B | ContextN/A | Max outputN/A | Open weightsNo | LicenseApache 2.0 |
Shieldstral 1.0 vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Shieldstral 1.0
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
MA
Mistral NeMo
Mistral AI
MA
Pixtral 12B
Mistral AI
MA
Ministral 8B
Mistral AI
MA
Codestral 22B
Mistral AI
MA
Mistral Large 2
Mistral AI
MA
Mistral Small 3 24B
Mistral AI
What is Shieldstral 1.0?
Key information about Shieldstral 1.0 and its available data.
Shieldstral is a 3B open-weight multimodal safety classifier from Mistral. It frames content moderation as policy-adaptive yes/no question answering: plain-language policies are supplied at inference time, and the model returns a calibrated safety score from a single forward pass for text, image, or text+image content.
0. 25857); per-bench WildGuard/HarmBench/ToxicChat/BeaverTails/Aegis/VLGuard ids are not in the catalog.
Data as of 2026-08-17.
FAQ
Common questions about Shieldstral 1.0.
When was Shieldstral 1.0 released?
Shieldstral 1.0's default version was released on Aug 4, 2026.
How much does Shieldstral 1.0 cost?
No official standard PAYG price is currently available for Shieldstral 1.0.
Who created Shieldstral 1.0?
Shieldstral 1.0 was created by Mistral AI.
What is the context window for Shieldstral 1.0?
A context window is not available for the default version.
Is Shieldstral 1.0 open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Shieldstral 1.0?
No provider offering is currently linked to the default version.
What models should I compare Shieldstral 1.0 with?
Nearby ranked alternatives include Phi 3.5 vision, IBM Granite 4.0 Tiny, Llama 3.2 11B.