{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"openrouter-mistralai-mistral-large-2407","slug":"mistralai-mistral-large-2407","name":"Mistral Large 2407","type":"model","url":"https://openrouter.ai/models/mistralai~mistral-large-2407","page_url":"https://unfragile.ai/mistralai-mistral-large-2407","categories":["chatbots-assistants"],"tags":["mistralai","api-access","text"],"pricing":{"model":"paid","free":false,"starting_price":"$2.00e-6 per prompt token"},"status":"active","verified":false},"capabilities":[{"id":"openrouter-mistralai-mistral-large-2407__cap_0","uri":"capability://text.generation.language.multi.turn.conversational.reasoning.with.context.preservation","name":"multi-turn conversational reasoning with context preservation","description":"Maintains conversation state across multiple turns using a transformer-based architecture with attention mechanisms that track dialogue history. The model processes the full conversation context (user messages, assistant responses, and implicit reasoning state) through its 141B parameter transformer to generate contextually coherent replies. Unlike stateless APIs, this implementation preserves semantic relationships across turns without explicit memory management, enabling complex multi-step reasoning within a single conversation thread.","intents":["build a chatbot that understands conversation history without manual state management","create an assistant that can reference earlier parts of a conversation to answer follow-up questions","develop a dialogue system where the model tracks implicit context and reasoning across turns"],"best_for":["teams building conversational AI products with complex multi-turn interactions","developers creating customer support chatbots requiring context awareness","builders prototyping dialogue systems where conversation history is critical"],"limitations":["context window is finite (32K tokens) — very long conversations require summarization or pruning","no persistent memory across separate conversation sessions — each new conversation starts fresh","latency increases with conversation length due to full context reprocessing on each turn"],"requires":["API key from Mistral AI or OpenRouter","HTTP client capable of streaming responses","conversation history management on client side (array of messages with roles)"],"input_types":["text (natural language queries)","code snippets embedded in conversation","structured prompts with system instructions"],"output_types":["text (natural language responses)","code (when reasoning about programming)","structured reasoning traces"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_1","uri":"capability://code.generation.editing.code.generation.and.completion.with.language.agnostic.synthesis","name":"code generation and completion with language-agnostic synthesis","description":"Generates syntactically correct code across 40+ programming languages by learning language-specific patterns during pretraining on diverse code repositories. The model uses transformer attention to understand code structure, variable scope, and API conventions, then generates completions that respect language semantics without explicit AST parsing. Supports both inline completion (filling gaps in existing code) and full function/module generation from natural language specifications.","intents":["generate boilerplate code or function stubs from natural language descriptions","complete partial code snippets with context-aware suggestions","translate algorithms between programming languages","generate test cases or documentation from code samples"],"best_for":["developers using IDE integrations or code editors for real-time completion","teams automating code generation in CI/CD pipelines","polyglot teams working across multiple programming languages"],"limitations":["generated code may contain logical errors or inefficiencies — requires human review and testing","no access to project-specific libraries or internal APIs unless provided in context","context window limits prevent generating very large files (>8K tokens) without chunking","may hallucinate function names or APIs that don't exist in the target language"],"requires":["API key for Mistral AI or OpenRouter","code context (existing file content, imports, or function signatures) for best results","target language specification in prompt or system message"],"input_types":["code snippets (partial or complete)","natural language descriptions of desired functionality","code comments or docstrings","function signatures or type hints"],"output_types":["code (syntactically valid in target language)","code with inline comments","test cases or example usage"],"categories":["code-generation-editing","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_10","uri":"capability://planning.reasoning.mathematical.reasoning.and.symbolic.computation","name":"mathematical reasoning and symbolic computation","description":"Solves mathematical problems including algebra, calculus, geometry, and logic through learned mathematical reasoning patterns. The model can work through multi-step problems, show intermediate steps, and verify solutions. This is implemented through training on mathematical datasets and chain-of-thought reasoning that prioritizes step-by-step problem solving.","intents":["solve math problems with step-by-step explanations","verify mathematical proofs or derivations","generate practice problems or homework solutions","explain mathematical concepts or theorems"],"best_for":["educators creating math tutoring systems","students using AI for homework help and learning","developers building educational platforms","researchers verifying mathematical reasoning"],"limitations":["symbolic computation is limited — cannot perform exact symbolic algebra like Mathematica","complex proofs may contain errors — requires verification by domain experts","numerical precision is limited by floating-point representation","very large numbers or complex expressions may exceed token limits"],"requires":["API key for Mistral AI or OpenRouter","mathematical problem in text or LaTeX format","optional: specification of desired solution format or level of detail"],"input_types":["mathematical problems in text or LaTeX","equations or expressions to solve","proofs to verify or explain"],"output_types":["step-by-step solutions","mathematical explanations","verified or corrected proofs"],"categories":["planning-reasoning","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_11","uri":"capability://code.generation.editing.code.review.and.debugging.with.architectural.analysis","name":"code review and debugging with architectural analysis","description":"Analyzes code for bugs, security issues, performance problems, and architectural concerns by understanding code semantics and common vulnerability patterns. The model can identify issues across multiple files, suggest fixes, and explain the reasoning behind recommendations. This is implemented through training on code repositories, security datasets, and best practices, combined with attention mechanisms that track variable flow and function calls.","intents":["review pull requests for bugs, security issues, or style violations","debug code by analyzing error messages and suggesting fixes","identify performance bottlenecks or architectural issues","suggest refactoring or code improvements"],"best_for":["development teams automating code review processes","security teams scanning code for vulnerabilities","developers debugging complex issues","teams enforcing code quality standards"],"limitations":["may miss subtle bugs or security issues — requires human review for critical code","architectural analysis is limited to visible code — cannot understand runtime behavior","performance analysis is heuristic-based — requires profiling for accurate optimization","context window limits prevent analyzing very large codebases in one pass"],"requires":["API key for Mistral AI or OpenRouter","code content (file or snippet)","optional: error messages, test failures, or specific concerns to focus on"],"input_types":["code snippets or files","error messages or stack traces","test failures or bug descriptions"],"output_types":["bug reports with explanations","suggested fixes or refactoring","security vulnerability reports","performance improvement recommendations"],"categories":["code-generation-editing","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_12","uri":"capability://text.generation.language.summarization.with.configurable.detail.levels.and.focus.areas","name":"summarization with configurable detail levels and focus areas","description":"Condenses long documents into summaries of varying lengths and focuses, preserving key information while removing redundancy. The model can generate executive summaries, detailed summaries, or summaries focused on specific topics by learning to identify important information and compress it. This is implemented through attention mechanisms that weight important tokens higher and training on summarization datasets.","intents":["create executive summaries of long documents for quick understanding","generate detailed summaries that preserve important context","extract key points from articles, papers, or reports","create topic-focused summaries highlighting specific aspects"],"best_for":["teams automating document processing and knowledge extraction","researchers quickly understanding large volumes of papers","business users summarizing reports or meeting notes","content platforms generating summaries for users"],"limitations":["summaries may omit important details — requires verification for critical applications","summary length is approximate — difficult to enforce exact token or word counts","topic-focused summaries may miss important context outside the focus area","very long documents may lose important details due to compression"],"requires":["API key for Mistral AI or OpenRouter","document content to summarize","optional: desired summary length or focus areas"],"input_types":["long-form documents","articles or papers","meeting notes or transcripts"],"output_types":["summaries of varying lengths","key points or bullet lists","topic-focused summaries"],"categories":["text-generation-language","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_13","uri":"capability://text.generation.language.sentiment.analysis.and.opinion.extraction.from.text","name":"sentiment analysis and opinion extraction from text","description":"Identifies sentiment (positive, negative, neutral) and extracts opinions, emotions, or attitudes from text by learning sentiment patterns and linguistic markers. The model can provide fine-grained sentiment analysis (aspect-based sentiment, emotion classification) and explain the reasoning behind sentiment judgments. This is implemented through training on sentiment datasets and attention mechanisms that identify sentiment-bearing tokens.","intents":["analyze customer reviews or feedback to understand satisfaction","monitor social media sentiment about products or brands","extract opinions from survey responses or interviews","classify emotions or attitudes in user-generated content"],"best_for":["teams analyzing customer feedback or reviews","social media monitoring and brand reputation management","market research and consumer insights","customer service teams prioritizing feedback"],"limitations":["sentiment analysis is context-dependent — sarcasm or irony may be misclassified","aspect-based sentiment requires explicit aspect specification — cannot auto-discover aspects","cultural or domain-specific sentiment markers may not be recognized","mixed sentiment (both positive and negative) is difficult to classify accurately"],"requires":["API key for Mistral AI or OpenRouter","text content to analyze","optional: specific aspects or emotions to focus on"],"input_types":["customer reviews or feedback","social media posts or comments","survey responses or interviews","product descriptions or marketing copy"],"output_types":["sentiment labels (positive, negative, neutral)","sentiment scores or confidence levels","emotion classifications","opinion summaries or key phrases"],"categories":["text-generation-language","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_2","uri":"capability://data.processing.analysis.structured.output.generation.with.json.schema.validation","name":"structured output generation with json schema validation","description":"Generates valid JSON and structured data by constraining the output space to match provided schemas or format specifications. The model uses guided decoding (token-level constraints during generation) to ensure output conforms to specified JSON schemas, XML structures, or other formal formats. This prevents hallucinated fields, enforces type correctness, and guarantees parseable output without post-processing validation.","intents":["extract structured data from unstructured text with guaranteed valid JSON output","generate API responses that conform to OpenAPI schemas","create configuration files or data models that must match specific formats","build data pipelines where downstream systems require strict schema compliance"],"best_for":["teams building data extraction pipelines requiring 100% valid output","developers creating API wrappers around LLMs with strict response contracts","data engineers automating ETL processes with schema-validated outputs"],"limitations":["schema constraints may limit expressiveness — very complex nested structures can reduce generation quality","guided decoding adds ~50-100ms latency per request due to token-level constraint checking","schema must be provided upfront — cannot dynamically infer structure from examples alone","deeply nested or recursive schemas may cause token budget exhaustion"],"requires":["API key for Mistral AI or OpenRouter","JSON schema definition (JSON Schema, OpenAPI, or similar format specification)","client library or API wrapper that supports schema-constrained generation"],"input_types":["unstructured text to extract from","JSON schema or format specification","natural language instructions for extraction"],"output_types":["valid JSON matching provided schema","structured data (XML, YAML if supported)","typed objects or dataclass instances"],"categories":["data-processing-analysis","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_3","uri":"capability://planning.reasoning.reasoning.focused.problem.decomposition.and.chain.of.thought","name":"reasoning-focused problem decomposition and chain-of-thought","description":"Decomposes complex problems into intermediate reasoning steps using learned patterns from chain-of-thought training data. The model generates explicit reasoning traces (showing work, considering alternatives, validating assumptions) before producing final answers. This is implemented through attention patterns that prioritize reasoning tokens and training objectives that reward step-by-step problem solving over direct answers.","intents":["solve multi-step math or logic problems with visible reasoning","debug code by having the model explain its reasoning about the bug","evaluate complex arguments or proposals by breaking them into components","generate detailed explanations for technical decisions or architectural choices"],"best_for":["educators building tutoring systems that need to show working","teams debugging complex systems where reasoning transparency is critical","developers building AI agents that need interpretable decision-making"],"limitations":["reasoning traces consume significant token budget — may exceed context limits on very complex problems","longer reasoning doesn't always correlate with better answers — can produce verbose but incorrect reasoning","reasoning format is natural language, not machine-parseable — requires post-processing to extract structured logic","latency increases 2-3x compared to direct answers due to intermediate step generation"],"requires":["API key for Mistral AI or OpenRouter","prompt engineering to explicitly request reasoning (e.g., 'think step by step')","sufficient context window to accommodate reasoning traces (16K+ tokens recommended)"],"input_types":["complex problems (math, logic, code debugging)","prompts with explicit reasoning requests","multi-part questions requiring decomposition"],"output_types":["reasoning traces (natural language steps)","final answers with justification","alternative approaches or considerations"],"categories":["planning-reasoning","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_4","uri":"capability://tool.use.integration.function.calling.and.tool.use.with.schema.based.dispatch","name":"function calling and tool use with schema-based dispatch","description":"Enables the model to decide when and how to call external functions or APIs by generating structured function calls based on provided tool schemas. The model receives a list of available functions (with parameters, descriptions, and types), reasons about which function to call, and generates properly formatted function calls (typically JSON) that client code can execute. This is implemented through training on function-calling datasets and constrained decoding to ensure valid function signatures.","intents":["build AI agents that can call APIs, databases, or custom functions to accomplish tasks","create assistants that decide when to search the web, check a database, or perform calculations","automate workflows where the model orchestrates multiple tool calls in sequence","enable the model to take actions in external systems based on user requests"],"best_for":["teams building autonomous agents or workflow automation systems","developers creating AI assistants that integrate with external APIs or services","builders implementing multi-step workflows where the model decides tool usage"],"limitations":["model cannot guarantee correct function selection — may call wrong function or misinterpret parameters","requires explicit function schema definitions — cannot auto-discover available tools","no built-in error handling or retry logic — client must handle function execution failures","function calling adds latency due to schema parsing and validation overhead"],"requires":["API key for Mistral AI or OpenRouter","function schema definitions (JSON Schema format or similar)","client code to execute returned function calls and feed results back to model","tool/function registry with clear descriptions and parameter specifications"],"input_types":["natural language requests or instructions","function schema definitions","context about available tools"],"output_types":["function calls (JSON with function name and parameters)","reasoning about which function to call","final response after function execution"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_5","uri":"capability://text.generation.language.multilingual.text.generation.and.translation.with.cross.lingual.reasoning","name":"multilingual text generation and translation with cross-lingual reasoning","description":"Generates coherent text in 50+ languages and translates between language pairs by learning cross-lingual representations during pretraining. The model understands semantic equivalence across languages and can reason about concepts in one language while generating in another. This is implemented through multilingual token embeddings and attention patterns that bridge language-specific syntax to shared semantic space.","intents":["translate content between languages while preserving meaning and tone","generate marketing copy or documentation in multiple languages from a single prompt","build chatbots that serve users in their preferred language","analyze sentiment or extract information from multilingual documents"],"best_for":["global teams building products for international markets","content creators localizing materials across multiple languages","companies providing customer support in multiple languages"],"limitations":["translation quality varies by language pair — low-resource languages may have lower quality","cultural nuances and idioms may not translate perfectly — requires human review for sensitive content","code-switching (mixing languages) may confuse the model in some contexts","no specialized domain knowledge for technical or legal translation without fine-tuning"],"requires":["API key for Mistral AI or OpenRouter","language specification in prompt (e.g., 'respond in French')","source text in supported language"],"input_types":["text in any supported language","language pair specification for translation","multilingual prompts or instructions"],"output_types":["text in target language","translated content preserving formatting","multilingual responses"],"categories":["text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_6","uri":"capability://text.generation.language.long.context.document.analysis.with.32k.token.window","name":"long-context document analysis with 32k token window","description":"Processes and analyzes documents up to 32,000 tokens (~24,000 words) in a single request by maintaining full context through the transformer's attention mechanism. The model can read entire documents, books, codebases, or conversation histories without summarization or chunking, enabling analysis that requires understanding relationships across distant parts of the document. This is implemented through optimized attention patterns and efficient memory usage in the 141B parameter model.","intents":["analyze entire research papers, legal documents, or technical specifications in one pass","summarize long documents while preserving key details and relationships","answer questions about specific sections while maintaining understanding of the full document","review entire codebases or pull requests for issues or improvements"],"best_for":["legal teams reviewing contracts or compliance documents","researchers analyzing papers or technical documentation","developers reviewing large codebases or pull requests","content teams summarizing or analyzing long-form content"],"limitations":["32K token limit still insufficient for very large documents (e.g., entire books) — requires chunking","latency increases with document length — processing 32K tokens takes ~5-10 seconds","attention complexity is O(n²) — very long contexts may cause memory issues on some hardware","model may lose focus on important details in very long documents due to attention dilution"],"requires":["API key for Mistral AI or OpenRouter","document content in text format (plain text, markdown, code, etc.)","sufficient API rate limits for large requests"],"input_types":["long-form text documents","code files or repositories","conversation histories","structured data (JSON, CSV) embedded in text"],"output_types":["summaries of long documents","answers to questions about document content","analysis or insights from full document context","extracted information or structured data"],"categories":["text-generation-language","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_7","uri":"capability://text.generation.language.instruction.following.and.task.specific.prompt.adaptation","name":"instruction-following and task-specific prompt adaptation","description":"Follows complex, multi-part instructions and adapts behavior based on system prompts and task specifications. The model learns to parse instruction hierarchies, prioritize conflicting directives, and maintain consistency with specified constraints throughout generation. This is implemented through instruction-tuning on diverse task datasets and training objectives that reward instruction adherence.","intents":["create specialized assistants with specific personalities, constraints, or expertise areas","automate content generation with precise style, tone, or format requirements","build systems that enforce business rules or compliance requirements through prompts","develop task-specific workflows where the model adapts to different instructions per request"],"best_for":["teams building customized AI assistants for specific use cases","content creators automating writing with consistent style and tone","companies enforcing compliance or brand guidelines through AI systems","developers building prompt-based workflows with varying requirements"],"limitations":["instruction following is probabilistic — complex or conflicting instructions may not be perfectly followed","model may misinterpret ambiguous instructions or prioritize wrong directives","very long or detailed instructions consume token budget, reducing space for actual task content","instruction adherence may degrade under adversarial or unusual prompts"],"requires":["API key for Mistral AI or OpenRouter","well-structured system prompts or instructions","clear specification of desired behavior, constraints, and output format"],"input_types":["system prompts with instructions","task-specific constraints or guidelines","user queries or content to process"],"output_types":["responses following specified instructions","content in specified format or style","outputs respecting specified constraints"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_8","uri":"capability://text.generation.language.knowledge.grounded.response.generation.with.factual.accuracy","name":"knowledge-grounded response generation with factual accuracy","description":"Generates responses grounded in training data knowledge while acknowledging uncertainty about information outside its training cutoff (April 2024). The model uses learned patterns to distinguish between high-confidence factual statements and speculative reasoning, and can indicate when information is uncertain or requires external verification. This is implemented through training objectives that reward factual accuracy and uncertainty quantification.","intents":["answer factual questions about historical events, scientific concepts, or technical topics","generate responses that acknowledge knowledge limitations and suggest verification","build systems that provide reliable information while flagging uncertain claims","create assistants that distinguish between facts and opinions"],"best_for":["teams building Q&A systems or knowledge bases","educators creating educational content with factual accuracy","companies providing customer information that must be reliable","developers building systems where factual errors have consequences"],"limitations":["knowledge cutoff is April 2024 — cannot answer questions about recent events","may hallucinate facts or confidently state incorrect information — requires fact-checking","no access to real-time information, current prices, or live data without external integration","factual accuracy varies by domain — better on well-represented topics in training data"],"requires":["API key for Mistral AI or OpenRouter","awareness of knowledge cutoff date (April 2024)","fact-checking or verification system for critical applications"],"input_types":["factual questions","requests for explanations or definitions","queries about historical or scientific topics"],"output_types":["factual answers with confidence indicators","explanations with source or reasoning","acknowledgment of uncertainty or knowledge gaps"],"categories":["text-generation-language","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"openrouter-mistralai-mistral-large-2407__cap_9","uri":"capability://text.generation.language.creative.writing.and.content.generation.with.style.control","name":"creative writing and content generation with style control","description":"Generates creative content (stories, poetry, marketing copy, dialogue) with controllable style, tone, and narrative elements. The model learns stylistic patterns from training data and can adapt to specified genres, voices, or writing styles through prompt engineering. This is implemented through attention mechanisms that capture stylistic features and training on diverse creative writing datasets.","intents":["generate marketing copy or product descriptions with specific tone and messaging","create story outlines, dialogue, or narrative content for games or entertainment","write poetry or creative text in specified styles or genres","generate multiple variations of content with different tones or approaches"],"best_for":["content creators and copywriters automating content generation","game developers generating dialogue and narrative content","marketing teams creating variations of copy for A/B testing","creative professionals using AI as a brainstorming tool"],"limitations":["creative output is non-deterministic — same prompt produces different results each time","style control through prompts is imprecise — may not perfectly match specified tone","generated content may be derivative or lack originality — requires human editing","very specific stylistic requirements may require multiple iterations to achieve"],"requires":["API key for Mistral AI or OpenRouter","clear specification of desired style, tone, genre, or voice","seed content or examples for style reference (optional but helpful)"],"input_types":["prompts specifying content type and style","example content for style reference","constraints or guidelines (length, tone, target audience)"],"output_types":["creative text (stories, poetry, copy)","dialogue or narrative content","multiple variations of content"],"categories":["text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":25,"verified":false,"data_access_risk":"low","permissions":["API key from Mistral AI or OpenRouter","HTTP client capable of streaming responses","conversation history management on client side (array of messages with roles)","API key for Mistral AI or OpenRouter","code context (existing file content, imports, or function signatures) for best results","target language specification in prompt or system message","mathematical problem in text or LaTeX format","optional: specification of desired solution format or level of detail","code content (file or snippet)","optional: error messages, test failures, or specific concerns to focus on"],"failure_modes":["context window is finite (32K tokens) — very long conversations require summarization or pruning","no persistent memory across separate conversation sessions — each new conversation starts fresh","latency increases with conversation length due to full context reprocessing on each turn","generated code may contain logical errors or inefficiencies — requires human review and testing","no access to project-specific libraries or internal APIs unless provided in context","context window limits prevent generating very large files (>8K tokens) without chunking","may hallucinate function names or APIs that don't exist in the target language","symbolic computation is limited — cannot perform exact symbolic algebra like Mathematica","complex proofs may contain errors — requires verification by domain experts","numerical precision is limited by floating-point representation","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.5,"ecosystem":0.24,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.35,"quality":0.2,"ecosystem":0.1,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:24.484Z","last_scraped_at":"2026-05-03T15:20:45.776Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=mistralai-mistral-large-2407","compare_url":"https://unfragile.ai/compare?artifact=mistralai-mistral-large-2407"}},"signature":"a26sr41O/1qo1WMdhjjQQVD5UmUSSc+/cKrq3PV8xHkKlZuKvnOoKtd/aTLlAF4RGYKT3iRMsPmoKGSFhTp0Bw==","signedAt":"2026-06-21T15:56:04.550Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/mistralai-mistral-large-2407","artifact":"https://unfragile.ai/mistralai-mistral-large-2407","verify":"https://unfragile.ai/api/v1/verify?slug=mistralai-mistral-large-2407","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}