完善了简体中文翻译

[sunny15007]

键名 英文原文 中文翻译
actions/clearLastUsedModel Clear last used model 清除上次使用的模型 actions/clearLastUsedModel/error Failed to clear last used model. 清除上次使用的模型失败。 style/viewMode/markdown Markdown Markdown
llm.load.promptTemplate/title Chat Template 聊天模板
llm.load.promptTemplate/subTitle The template used by engine protocol for chat, tools, reasoning, and multimodal formatting. Changing this requires reloading the model. 引擎协议用于聊天、工具、推理和多模态格式化的模板。更改此设置需要重新加载模型。 llm.load.numParallelSessions/title Max Concurrent Predictions 最大并发预测数 llm.load.numParallelSessions/subTitle Maximum number of predictions the model can run at a given time. The speed of each individual prediction may decrease with concurrency, but each prediction will start faster and higher total throughput can be achieved 模型在给定时间可同时运行的最大预测数。随着并发数增加，单个预测速度可能下降，但每个预测启动会更快，且能实现更高的总吞吐量。 llm.load.useUnifiedKvCache/title Unified KV Cache 统一 KV 缓存 llm.load.useUnifiedKvCache/subTitle Controls whether concurrent predictions share a single KV cache, saving memory. Disabling this ensures each prediction can utilize the full context length, at the cost of using more memory 控制并发预测是否共享单个 KV 缓存以节省内存。禁用此选项可确保每个预测都能利用完整的上下文长度，但会消耗更多内存。 llm.load.numCpuExpertLayersRatio/title Number of layers for which to force MoE weights onto CPU 强制将 MoE 权重置于 CPU 的层数 llm.load.numCpuExpertLayersRatio/subTitle Number of layers to force the experts into CPU. Saves VRAM and can be faster than partial GPU offload. Not recommended if the model fits entirely in VRAM. 强制将专家层置于 CPU 的层数。可节省显存，且可能比部分 GPU 卸载更快。如果模型完全适合显存，则不建议使用。 llm.load.numCpuExpertLayersRatio/info Specifies the number of layers to force the expert into CPU. Leaves attention layers on GPU, saving VRAM while keeping inference fairly fast. 指定强制将专家层置于 CPU 的层数。将注意力层保留在 GPU 上，节省显存的同时保持推理速度。 llm.load.llama.physicalBatchSize/subTitle Maximum number of prompt tokens to process in one chunk. Larger values may improve performance, but can use more memory 每个块处理的最大提示词 token 数。较大的值可能提升性能，但会占用更多内存。 serverOptions/allowMcp/title Allow Remote MCP 允许远程 MCP serverOptions/allowMcp/hint Allow using MCPs that are not in your mcp.json. These MCP connections are ephemeral, only existing as long as the request. At the moment, only remote MCPs are supported. 允许使用不在 mcp.json 中的 MCP。这些 MCP 连接是临时的，仅在请求期间存在。目前仅支持远程 MCP。 serverOptions/allowMcp/mode/off Off 关闭
serverOptions/allowMcp/mode/off/hint Don't allow server requests to use MCP 不允许服务器请求使用 MCP。 serverOptions/allowMcp/mode/remote Remote 远程
serverOptions/allowMcp/mode/remote/hint Allow connection to remote MCP servers 允许连接到远程 MCP 服务器。 serverOptions/start/error Failed to start the server 启动服务器失败 serverOptions/stop/error Failed to stop the server 停止服务器失败 endpoints/openaiCompatRest/segmentedLabel OpenAI-like 类似 OpenAI 的接口 endpoints/openaiCompatRest/postResponses Advanced interface for generating model responses. Create stateful interactions by passing the id of a previous responses as input to the next. 用于生成模型响应的高级接口。通过将先前响应的 id 作为下一次请求的输入，可创建有状态的交互。 endpoints/lmStudioRest/segmentedLabel LM Studio LM Studio endpoints/lmStudioRestV1/getModels List available models 列出可用模型 endpoints/lmStudioRestV1/postModelsLoad Load a model with options 使用选项加载模型 endpoints/lmStudioRestV1/postModelsDownload Download a model 下载模型 endpoints/lmStudioRestV1/postChat Chat with a model. Supports stateful multi-turn conversations and MCP 与模型聊天，支持有状态多轮对话和 MCP endpoints/lmStudioRestV1/getModelsDownloadStatus Get the status of a model download 获取模型下载状态 endpoints/anthropicCompatRest/segmentedLabel Anthropic-compatible 兼容 Anthropic 的接口 indexingPageLoaderText Indexing models... 正在索引模型... loader.guardrails.estimatedMemoryUsage Estimated Memory Usage 预估内存使用量 loader.guardrails.total Total 总计
loader.guardrails.gpu GPU GPU
loader.guardrails.unavailable Memory estimate unavailable for this model 此模型的预估内存不可用 loader.guardrails.notEnoughResources Not enough resources to load the model with the current settings 当前设置下资源不足以加载此模型 loader.guardrails.notEnoughResources/options Options 选项 loader.guardrails.notEnoughResources.moreInfoSection.appearsNotEnoughMemory It appears your system does not have enough memory to load this model. 系统似乎没有足够内存加载此模型。 loader.guardrails.notEnoughResources.moreInfoSection.ifYouBelieveThisIsIncorrect You can adjust the model loading guardrails in settings or hold to load anyway. 您可以在设置中调整模型加载保护，或按住  强制加载。 loader.guardrails.notEnoughResources.moreInfoSection.warning Loading a model that is too large may overload your system and cause it to freeze. 加载过大的模型可能导致系统过载并卡死。 loader.guardrails.notEnoughResources.alwaysAllowLoadAnyway (Not recommended) Always allow 'Load Anyway' without holding Alt/Option （不推荐）始终允许不按 Alt/Option 直接“强制加载” appNavigationBarPositionLabel Navigation Bar position 导航栏位置 appNavigationBarPositionTop Top 顶部
appNavigationBarPositionLeft Left 左侧
modelDefaultsLabel Model Defaults 模型默认值
modelLoadingGuardrails.alwaysAllowLoadAnyway (Not recommended) Always allow 'Load Anyway' without holding Alt/Option （不重复，同上一条） defaultContextLength {"label": "Default Context Length", "maxTitle": "Model Maximum", "customTitle": "Custom value", "maxSubtitle": "Use the maximum context length supported by each model.", "customSubtitle": "Set the default context length for loading new models. If the model's supported maximum context length is lower, that value will be used.", "invalidNaNError": "Invalid context length value. Using {{value}}", "invalidRangeError": "Invalid context length value. Should be in the range of 1 and 2^30. Using {{value}}", "largeContextWarning": "The higher the context length, the more memory the model will take. If you are unsure, don't change the default"} {"label": "默认上下文长度", "maxTitle": "模型最大值", "customTitle": "自定义值", "maxSubtitle": "使用每个模型支持的最大上下文长度。", "customSubtitle": "设置加载新模型时的默认上下文长度。如果模型支持的最大上下文长度更小，则使用该值。", "invalidNaNError": "无效的上下文长度值。使用 {{value}}", "invalidRangeError": "无效的上下文长度值。应在 1 到 2^30 之间。使用 {{value}}", "largeContextWarning": "上下文长度越高，模型占用内存越多。如果不确定，请不要更改默认值。"} jitTTL {"subtitle": "JIT-loaded models will be automatically unloaded after being idle for the specified duration."} {"subtitle": "即时加载的模型在闲置指定时长后将自动卸载。"} beta Beta 测试版

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[sunny15007]

I have read the CLA Document and I hereby sign the CLA