MoonshotAI: Kimi K3.1 Pro

{"about":"Kimi K3.1 Pro est un modèle de Moonshot AI, le premier de la série K3.1. Il est fondé sur Kimi K3, avec une augmentation importante du volume de post-entraînement et un apprentissage par renforcement conjoint sur de nombreux domaines. La présentation officielle met en avant la programmation, les projets en trois dimensions, la recherche financière et les tâches nécessitant plusieurs étapes liées entre elles.\n\n### Programmation et projets techniques\n\nLe modèle peut servir à analyser du code existant, à préparer des modifications et à planifier le développement. Fournissez l’objectif, l’implémentation actuelle et les contraintes : par exemple, les interfaces à préserver et les critères d’un résultat correct. Vous pouvez demander une explication, le texte d’un correctif, des scénarios de test ou une séquence de mise en œuvre.\n\nPour les jeux, les scènes interactives et la modélisation paramétrique, la préparation de scripts et de logique technique constitue un usage pertinent. La documentation du concepteur évoque le travail avec Blender, la CAO et les moteurs de jeu. Dans Ailoot, cette fiche produit du texte : il est donc pratique de formuler la demande comme du code pour l’outil choisi, une description géométrique ou un plan de construction de scène, à exécuter ensuite dans votre propre environnement.\n\n### Analyse des données fournies\n\nPour des documents, vous pouvez demander une comparaison des sources, une explication d’un calcul, une vérification de la cohérence des hypothèses ou une note de recherche. Pour un modèle financier, fournissez les données de départ, les unités de mesure et les relations entre les indicateurs. Le modèle peut proposer des formules sous forme textuelle et une démarche de vérification ; pour un rapport, il peut préparer une structure et une argumentation liées aux contenus fournis.\n\n### Comment obtenir une réponse utile\n\n1. Indiquez la tâche et le résultat attendu : explication, code, plan ou conclusion analytique.\n2. Joignez les textes nécessaires et précisez quelles sources ont priorité.\n3. Décrivez les critères de vérification : conditions à préserver, format, précision des calculs et sections obligatoires.\n4. Poursuivez par des précisions, en conservant les décisions importantes et les contraintes initiales.\n\n### Entrée et résultat dans Ailoot\n\nLa fiche accepte du texte et renvoie du texte. Les contenus adaptés comprennent des instructions, des extraits de documents, du code source, des tableaux au format texte et l’historique des échanges. Sur K3.1 Pro, le raisonnement est toujours activé ; la documentation du modèle prévoit les niveaux low, high et max. Dans Standard, la limite d’entrée est de 262 144 tokens et celle de la réponse de 65 536 tokens. Pour une requête volumineuse, tenez compte de l’historique de la conversation et adaptez la quantité de contenu à la question précise.","alias_of":null,"aliases":["k3.1-pro","kimi-k3.1-pro"],"author":"MoonshotAI","canonical_model":"moonshotai/kimi-k3.1-pro","ctx":262144,"formats":["chat","gemini","messages","responses"],"id":"moonshotai/kimi-k3.1-pro","input":["text"],"kind":"text","max_out":65536,"methods":[{"cache_read":0.2,"cache_write":2.0,"cache_write_1h":null,"cache_write_5m":2.0,"conditional_meter":null,"ctx":262144,"family":"anthropic","fast_multiplier":null,"flex_multiplier":null,"fmt":"messages","http":"POST","img_in":null,"img_out":null,"in":2.0,"input":["text"],"label":"Anthropic Messages","max_out":65536,"max_tool_calls":null,"out":10.0,"output":["text"],"path":"/v1/messages","request_model_id":"k3.1-pro","route_label":"Core","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.2,"cache_write":2.0,"cache_write_1h":null,"cache_write_5m":2.0,"conditional_meter":null,"ctx":262144,"family":"openai","fast_multiplier":null,"flex_multiplier":null,"fmt":"responses","http":"POST","img_in":null,"img_out":null,"in":2.0,"input":["text"],"label":"OpenAI Responses","max_out":65536,"max_tool_calls":null,"out":10.0,"output":["text"],"path":"/v1/responses","request_model_id":"k3.1-pro","route_label":"Core","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.2,"cache_write":2.0,"cache_write_1h":null,"cache_write_5m":2.0,"conditional_meter":null,"ctx":262144,"family":"openai","fast_multiplier":null,"flex_multiplier":null,"fmt":"chat","http":"POST","img_in":null,"img_out":null,"in":2.0,"input":["text"],"label":"OpenAI Chat","max_out":65536,"max_tool_calls":null,"out":10.0,"output":["text"],"path":"/v1/chat/completions","request_model_id":"k3.1-pro","route_label":"Core","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.2,"cache_write":2.0,"cache_write_1h":null,"cache_write_5m":2.0,"conditional_meter":null,"ctx":262144,"family":"gemini","fast_multiplier":null,"flex_multiplier":null,"fmt":"gemini","http":"POST","img_in":null,"img_out":null,"in":2.0,"input":["text"],"label":"Gemini generateContent","max_out":65536,"max_tool_calls":null,"out":10.0,"output":["text"],"path":"/v1beta/models/{model}:generateContent","request_model_id":"k3.1-pro","route_label":"Core","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":4.0,"cache_write_1h":null,"cache_write_5m":4.0,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"chat","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"OpenAI Chat","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1/chat/completions","request_model_id":"kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":4.0,"cache_write_1h":null,"cache_write_5m":4.0,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"responses","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"OpenAI Responses","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1/responses","request_model_id":"kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":4.0,"cache_write_1h":null,"cache_write_5m":4.0,"conditional_meter":null,"ctx":null,"family":"gemini","fast_multiplier":1,"flex_multiplier":1,"fmt":"gemini","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"Gemini generateContent","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1beta/models/kimi-k3.1-pro:generateContent","request_model_id":"kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"anthropic","fast_multiplier":1,"flex_multiplier":1,"fmt":"messages","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"Anthropic Messages","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1/messages","request_model_id":"moonshotai/kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"responses","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"OpenAI Responses","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1/responses","request_model_id":"moonshotai/kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"openai","fast_multiplier":1,"flex_multiplier":1,"fmt":"chat","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"OpenAI Chat","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1/chat/completions","request_model_id":"moonshotai/kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]},{"cache_read":0.4,"cache_write":null,"cache_write_1h":null,"cache_write_5m":null,"conditional_meter":null,"ctx":null,"family":"gemini","fast_multiplier":1,"flex_multiplier":1,"fmt":"gemini","http":"POST","img_in":null,"img_out":null,"in":4.0,"input":["text"],"label":"Gemini generateContent","max_out":null,"max_tool_calls":null,"out":20.0,"output":["text"],"path":"/v1beta/models/moonshotai/kimi-k3.1-pro:generateContent","request_model_id":"moonshotai/kimi-k3.1-pro","route_label":"Ultra","status_reads":[],"tiers":[],"tool_prices":null,"unit":"token","units":[]}],"model_aliases":[{"canonical_model_id":"moonshotai/kimi-k3.1-pro","request_model_id":"k3.1-pro","route_label":"Core"},{"canonical_model_id":"moonshotai/kimi-k3.1-pro","request_model_id":"kimi-k3.1-pro","route_label":"Ultra"},{"canonical_model_id":"moonshotai/kimi-k3.1-pro","request_model_id":"moonshotai/kimi-k3.1-pro","route_label":"Ultra"}],"name":"MoonshotAI: Kimi K3.1 Pro","price":{"in":2.0,"in_label":null,"out":10.0,"out_label":null,"sort_in":2.0,"sort_out":10.0,"varies":true},"public_id":"moonshotai/kimi-k3.1-pro","short":"Kimi K3.1 Pro aide à analyser des contenus complexes, à écrire du code et à préparer des solutions textuelles pour les projets, la modélisation et l’analyse."}