LLM Comparison
Leanstral 1.5 vs Llama Guard 3 8B
Side-by-side specs, pricing & capabilities
Add to comparison
2/6 models| Organization | ||
| OpenTools Score | ||
| Family | Leanstral | Llama |
| Status | Current | Current |
| Release Date | Jun 2026 | Feb 2025 |
| Context Window | 256K tokens | 131K tokens |
| Input Price | Free | $0.48/M tokens |
| Output Price | Free | $0.03/M tokens |
| Pricing Notes | Mistral model card lists Leanstral 1.5 price as $0. Changelog says the Labs endpoint is scheduled to retire on 2026-09-30. | — |
| Capabilities | textreasoningcodingformal-verificationtool-usestructured-output | textcode |
| Max Output | 128K tokens | 100 tokens |
| API Identifier | labs-leanstral-1-5 | meta-llama/llama-guard-3-8b |
| Benchmarks | ||
| LG3 code-interpreter safety F1 | — | 88.5%meta |
| LG3 English safety AUPRC | — | 98.5%meta |
| LG3 English safety F1 | — | 93.9%meta |
| LG3 tool-use safety F1 | — | 85.6%meta |
| View Leanstral 1.5 | View Llama Guard 3 8B | |
Cost Calculator
Enter your expected monthly token usage to compare costs.
| Model | Input | Output | Total / mo | vs Best |
|---|---|---|---|---|
| Leanstral 1.5Cheapest | $0.00 | $0.00 | $0.00 | — |
| Llama Guard 3 8B | $0.48 | $0.02 | $0.50 | +0% |
Compare hosting configurations
Compare two exact setups, including different configurations of the same model. Hardware fit, measured speed and costs are separate decisions.
Add your cost assumptions (optional)
Use hours for the same scenario period. Cloud billing includes idle time; active hours alone do not set the bill. Local electricity uses whole-system attributable power, powered hours and your tariff. Missing inputs remain unknown.
Choose two different configurations to see their tradeoffs.
Mistral AI
Leanstral 1.5
Leanstral 1.5 is Mistral AI free Lean 4 formal proof-engineering model for automated theorem proving, autoformalization, and real-world code verification. Official Mistral docs list it as a Labs model with 119B total parameters, 6.5B active parameters, a 256k context window, 128k max output, $0 pricing, and the labs-leanstral-1-5 API identifier.
Meta
Llama Guard 3 8B
Llama Guard 3 8B is a Meta safety-classification model for prompt and response moderation across English, multilingual, search, and code-interpreter use cases.
More Comparisons
Looking for more AI models?
Browse All LLMs