使用左右方向键逐个查看每次检测结果。
QuickSilver Pro 系统状态
状态未知
检查中...
- 3 项服务3 项服务: 检查中...
- 监测 91 个模型91 个模型: 检查中...
- 合成探测 · 每 3 分钟更新每 3 分钟探测
服务
模型可用性
- DeepSeek V4 Flash
deepseek-v4-flash检查中...official 0731 agent model, 1M context, Responses API
- DeepSeek V4.1 Flash
deepseek-v4.1-flash检查中...official DeepSeek V4.1 Flash — new architecture, faster, 1M context, Responses API
- DeepSeek V4.1 Flash 1017
deepseek-v4.1-flash-1017检查中...October 17 snapshot, 1M context, tools and optional reasoning
- DeepSeek V4 Pro
deepseek-v4-pro检查中...premium reasoning, 1M context
- Qwen3.8 Max
qwen3.8-max检查中...Qwen 3.8 flagship, 2.4T MoE, 1M context, autonomous coding
- Qwen3.8 Max Prime
qwen3.8-max-prime检查中...Qwen 3.8 top-tier flagship, deepest reasoning & autonomous coding, 1M context
- Qwen3.8 Omni Flash
qwen3.8-omni-flash检查中...Qwen's first agentic omni model: native image, audio and video understanding with tool use and reasoning, 1M context, at Alibaba's first-party price
- Qwen3.7 Max
qwen3.7-max检查中...Qwen 3.7 flagship, 1M context, thinks by default
- Qwen3.7 Plus
qwen3.7-plus检查中...Qwen 3.7 agent flagship, 1M context, long-horizon coding
- Qwen3.7 Flash
qwen3.7-flash检查中...fast multimodal agents, vision and tool use, 1M context
- Qwen3.6 Plus
qwen3.6-plus检查中...1T-MoE flagship, 1M context, thinks by default
- Qwen3.6-35B-A3B
qwen3.6-35b检查中...262K long-context, MoE, drop-in 3.5 upgrade
- Qwen3.8 27B
qwen3.8-27b检查中...Cloud-only Qwen 27B with a 1M-token context window
- Qwen3.8 27B Decentered
qwen3.8-27b-decentered检查中...Share Compute community Macs: 30% lower token prices; cloud fallback billed at Qwen3.8 27B prices
- Qwen3.8 Flash Next
qwen3.8-flash-next检查中...Qwen's open Qwen4-architecture preview: 6B-active MoE, fast and cheap, with a 1M-token context window
- Kimi K2.6
kimi-k2.6检查中...Opus-class reasoning, 256K context
- Kimi K2.7 Code
kimi-k2.7-code检查中...Agentic-coding K2, 256K context, coding-tuned K2.6 successor
- Kimi K3.1 Pro
kimi-k3.1-pro检查中...1M context, image input, tools and optional reasoning
- Kimi K3.1 Pro Highspeed
kimi-k3.1-pro-highspeed检查中...1M context, image input, tools and optional reasoning
- Kimi K3
kimi-k3检查中...2.8T multimodal reasoning flagship, 1M context
- Muse Wonder
muse-wonder检查中...Frontier reasoning for agentic personal and professional work, 1M context
- Muse Spark Open
muse-spark-open检查中...Meta open-weight model for coding and tools, 1M context
- Muse Spark 1.3
muse-spark-1.3检查中...Coding & agentic reasoning model, 1M context
- Muse Spark 1.2
muse-spark-1.2检查中...Coding-focused reasoning model, 1M context
- Muse Glimmer 30B
muse-glimmer-30b检查中...Compact agentic multimodal model, distilled from Muse Spark
- GLM 5.3
glm-5.3检查中...Z.ai latest reasoning flagship, 1M context, complex software engineering & long-horizon agents
- GLM 5.3 Prime
glm-5.3-prime检查中...Z.ai GLM 5.3 Prime, top reasoning flagship, 1M context, complex software engineering
- GLM 5.3 Flash
glm-5.3-flash检查中...Z.ai fast open-weights reasoning model, 1M context at workhorse pricing
- GLM 5.2
glm-5.2检查中...Z.ai reasoning flagship, 1M context, long-horizon agents & coding
- Nemotron 3 Ultra
nemotron-3-ultra检查中...550B-A55B open MoE for frontier reasoning and agent orchestration
- Nemotron 3.5 Lightning
nemotron-3.5-lightning检查中...30B-A3B open MoE for fast, tool-heavy agents
- GPT-OSS 120B
gpt-oss-120b检查中...OpenAI's open-weight 117B MoE, high-volume agents & reasoning, 131K context
- GPT-6 Astra
gpt-6-astra检查中...OpenAI's GPT-6 flagship, long-horizon agentic coding & deep research, 1M context
- GPT-6 Luna
gpt-6-luna检查中...OpenAI's cost-efficient GPT-6, high-volume chat & lightweight agentic work, 1M context
- GPT-6.1 Sol
gpt-6.1-sol检查中...OpenAI's GPT-6.1 Sol, near-Astra agentic coding at Sol pricing, 1M context
- GPT-6.1 Terra
gpt-6.1-terra检查中...OpenAI's GPT-6.1 Terra, mid-tier everyday agentic coding, writing, and screenshot understanding, 1M context
- GPT-6.1 Terra Pro
gpt-6.1-terra-pro检查中...GPT-6.1 Terra in pro reasoning mode -- same model, same per-token price, far more reasoning tokens per request on complex tasks
- GPT-6 Sol
gpt-6-sol检查中...OpenAI's GPT-6 mid-flagship, strong reasoning & agentic coding, 1M context
- GPT-5.6 Luna
gpt-5.6-luna检查中...OpenAI's fast, cost-efficient GPT-5.6 tier, 1M context
- GPT-5.6 Terra
gpt-5.6-terra检查中...OpenAI's balanced GPT-5.6 mid-tier, 1M context
- GPT-5.6 Sol
gpt-5.6-sol检查中...OpenAI's flagship GPT-5.6, complex reasoning & agentic coding, 1M context
- Grok 4.5
grok-4.5检查中...xAI's smartest model, frontier coding, knowledge work & STEM, 500K context
- Grok 4.6
grok-4.6检查中...frontier coding, knowledge work and STEM, vision, 500K context
- Grok 4.7
grok-4.7检查中...frontier coding, knowledge work and STEM, vision, 500K context
- MiniMax M3
minimax-m3检查中...MiniMax open-weight frontier model, 1M context, agentic coding
- MiMo-V2.6-Pro
mimo-v2.6-pro检查中...Xiaomi flagship open-weight MiMo-V2.6-Pro, top open model on AA Intelligence Index, coding & agentic, vision, 1M context
- MiMo-V2.6-Flash
mimo-v2.6-flash检查中...Xiaomi open-weight MiMo-V2.6-Flash, fast low-cost coding & agentic, vision, 1M context
- MiMo-V2.5
mimo-v2.5检查中...Xiaomi open-weight MiMo-V2.5, cost-efficient coding & agentic, 1M context
- Hy4 Preview
hy4-preview检查中...Tencent Hy4 (preview), reasoning & agentic coding, 1M context
- Hy3
hy3检查中...Tencent Hy3, selectable reasoning & agentic coding, 262K context
- Step 5 Preview
step-5-preview检查中...StepFun reasoning model for tools and image understanding, 1M context
- Mistral Large 4
mistral-large-4检查中...Mistral's frontier multimodal model for coding and agents, 524K context
- Jev 1.13
jev-1.13检查中...Typed decisions with calibrated probabilities from a single forward pass: routing, classification and scoring without text generation
- Trio-Spark v1.3
trio-spark-v1.3检查中...System-1 decision model: one forward pass, calibrated probabilities, no generated tokens
- Claude Opus 5.5
claude-opus-5-5检查中...Anthropic's most capable Claude: agentic coding, knowledge work & computer use, 1M context
- Claude Opus 5
claude-opus-5检查中...Anthropic Opus 5 for demanding reasoning, coding & long-horizon agentic work, 1M context
- Claude Fable 5.1
claude-fable-5-1检查中...Anthropic Mythos-class model for deep reasoning and long-horizon agentic work (5.1)
- Claude Fable 5
claude-fable-5检查中...Anthropic Mythos-class flagship, most capable reasoning
- Claude Opus 4.8
claude-opus-4-8检查中...Anthropic flagship, top-tier reasoning & agentic
- Claude Sonnet 5.5
claude-sonnet-5-5检查中...Anthropic's newest Sonnet, faster and fewer tokens per task, 1M context
- Claude Sonnet 5
claude-sonnet-5检查中...Anthropic's newest Sonnet tier, 1M context
- Claude Haiku 5.5
claude-haiku-5-5检查中...Anthropic's newest small, fast model for subagents and high-volume work, 1M context
- Claude Haiku 4.5
claude-haiku-4-5检查中...fast, low-cost Anthropic, high-volume
- Gemini 4 Argon
gemini-4-argon检查中...Google's new flagship reasoning model, 1M context; introductory price
- Gemini 3.7 Flash
gemini-3.7-flash检查中...fast multimodal Flash for agents; promotional pricing through 2026-12-31
- Gemini 3.8 Flash
gemini-3.8-flash检查中...most capable Flash for agents; promotional pricing through 2026-12-31
- Gemini 3.6 Flash
gemini-3.6-flash检查中...current general-purpose Flash GA, 1M context
- Gemini 3.5 Flash-Lite
gemini-3.5-flash-lite检查中...current low-cost Flash-Lite GA, 1M context
- Gemini 3.5 Flash
gemini-3.5-flash检查中...Google's next-gen Flash GA, 1M context, thinking
- Gemini 3.1 Pro Preview
gemini-3.1-pro-preview检查中...Google's flagship reasoning, 1M context, thinking
- Nano Banana Pro (Gemini 3 Pro Image)
gemini-3-pro-image检查中...Google's Nano Banana Pro at 1K, billed per image
- Nano Banana 2.1
gemini-nano-banana-2.1检查中...Google's newest Nano Banana at 1K, billed per image
- Gemini 3 Flash Preview
gemini-3-flash-preview检查中...legacy compatibility; migrate to Gemini 3.6 Flash
- Gemini 3.1 Flash Lite
gemini-3.1-flash-lite检查中...legacy compatibility; migrate to Gemini 3.5 Flash-Lite
- FLUX.2 Pro
flux.2-pro检查中...flagship image generation, billed per image
- FLUX.1 Schnell
flux.1-schnell检查中...ultra-fast open image generation, billed per image
- SDXL Turbo
sdxl-turbo检查中...fast open image generation (SDXL Turbo), billed per image
- FLUX.2 Klein
flux.2-klein检查中...open FLUX.2 image generation, billed per image
- Qwen Image 3.0
qwen-image-3.0检查中...Alibaba text-to-image, 512x512 to 2560x2560, flat $0.04 per image; optional prompt_extend
- Seedream 5.0 Pro
seedream-5.0-pro检查中...ByteDance's Seedream 5.0 Pro: flagship photorealistic image generation, billed per image
- Seedream 4
seedream-4检查中...ByteDance Seedream 4: fast, high-quality image generation, billed per image
- Bria FIBO 1.5
bria-fibo-1.5检查中...Bria FIBO 1.5: image generation trained on fully licensed data (commercially safe), billed per image
- Bria FIBO 2.0
bria-fibo-2.0检查中...Bria FIBO 2.0: Bria reports training on fully licensed data for commercially safe image generation, billed per image
- GPT Image 2
gpt-image-2检查中...OpenAI's GPT Image 2 at a fixed ~1.5MP standard output, billed per image
- GPT Image 2.5 Flare
gpt-image-2.5-flare检查中...OpenAI's GPT Image 2.5, speed tier, at medium quality, billed per image
- GPT Image 2.5 Sunburst
gpt-image-2.5-sunburst检查中...OpenAI's GPT Image 2.5, precision tier, at medium quality, billed per image
- Qwen3 ASR
qwen3-asr检查中...Speech to text
- Nemotron 3.5 ASR
nemotron-3.5-asr检查中...Speech to text
- Whisper Large v3 Turbo
whisper-large-v3-turbo检查中...Speech to text
- Chatterbox Turbo
chatterbox-turbo检查中...English text to speech
- Qwen3 TTS
qwen3-tts检查中...Chinese and English text to speech
基础设施路线图
当前基础
已在精选模型目录上线
每个模型都按其公布的单模型费率计费，每一项价格对比都可以对照 /pricing.json 核查。收窄的运营面让这些数字保持可验证，同一套接口也会直接延续到下一个里程碑。
下一个里程碑
在 H100/H200 上运行自有推理栈
使用专用 GPU 自托管推理服务，结合 SGLang + continuous batching、EAGLE-3 speculative decoding、通过 DeepGEMM 做 FP8 quantization，以及 SageAttention / ThunderMLA 自定义内核。到那时
system_fingerprint会稳定下来（只有我们升级推理栈时才变化），可重复 seed 工作流也会真正可用。目标：DeepSeek V4 系列的价格比当前再低 30-50%。长期方向
机柜托管数据中心 + AIDC 合作
从租用（Vast.ai）转向自有或托管机柜，并在合适的地方与 AI 数据中心运营商合作。目标是做出地球上最便宜且可靠的开源模型推理服务：从全栈到底层工程都由我们自己掌控。
关于此页面
后端和官网行显示从你的浏览器发出的连通性检测。API 行汇总后端报告的所列模型检测结果。 模型行反映的是我们后端每 3 分钟发起一次、真实 1-token 的服务端探测。 历史条展示的是近期探测结果，数据存储在当前浏览器的 localStorage；如果你切换设备，这些历史会被清空。
公开可用性追踪始于 2026-04-16. 如果你需要带合同约束的 SLA 和第三方监控历史，请 联系我们。