9i9inference

Status

Service Status

Uptime and response duration for each model, measured from real requests to the 9inference API.

Some models are experiencing an outage
Updated 22:50 WIB

Package API

Token package models via /v1/package.

  • claude-sonnet-5-5

    claude-sonnet-5-5
    Operational
    30 days agoToday
    30-day uptime
    91.68%
    Median response duration (24h)
    10.6 s

  • deepseek-v4.1-flash

    deepseek-v4.1-flash
    Operational
    30 days agoToday
    30-day uptime
    99.55%
    Median response duration (24h)
    3.5 s

  • deepseek-v4.1-flash-1017

    deepseek-v4.1-flash-1017
    Operational
    30 days agoToday
    30-day uptime
    99.96%
    Median response duration (24h)
    5.8 s

  • deepseek-v4.1-flash-uncensored

    deepseek-v4.1-flash-uncensored
    Operational
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    21.4 s

  • deepseek-v4-flash-vision-exp

    deepseek-v4-flash-vision-exp
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • deepseek-v4-pro-0813

    deepseek-v4-pro-0813
    Outage
    30 days agoToday
    30-day uptime
    90.37%
    Median response duration (24h)
    113.7 s

  • glm-5.3-flash

    glm-5.3-flash
    Operational
    30 days agoToday
    30-day uptime
    99.75%
    Median response duration (24h)
    3.2 s

  • kimi-k3

    kimi-k3
    Operational
    30 days agoToday
    30-day uptime
    99.88%
    Median response duration (24h)
    15.9 s

  • qwen3.8-max

    qwen3.8-max
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • DeepSeek V4 Flash

    deepseek-v4-flash
    Not enough data
    30 days agoToday
    30-day uptime
    —
    Median response duration (24h)
    —

  • DeepSeek V4 Flash 0731

    deepseek-v4-flash-0731
    Operational
    30 days agoToday
    30-day uptime
    99.98%
    Median response duration (24h)
    5.9 s

  • DeepSeek V4 Pro

    deepseek-v4-pro
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • GLM 5.1

    glm-5.1
    Not enough data
    30 days agoToday
    30-day uptime
    —
    Median response duration (24h)
    —

  • GLM 5.2

    glm-5.2
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • GLM 5.3

    glm-5.3
    Operational
    30 days agoToday
    30-day uptime
    99.99%
    Median response duration (24h)
    9.2 s

  • GPT OSS 120B

    gpt-oss-120b
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • Kimi K2.6

    kimi-k2.6
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • MiniMax M3

    minimax-m3
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • Nemotron 3 Ultra

    nemotron-3-ultra
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • Kimi K2.7 Code

    kimi-k2.7-code
    Not enough data
    30 days agoToday
    30-day uptime
    100.00%
    Median response duration (24h)
    —

  • Gemini 3.8 Flash High

    gemini-3.8-flash-high
    Operational
    30 days agoToday
    30-day uptime
    98.62%
    Median response duration (24h)
    6.1 s

PAYG API

Pay-per-token models via /v1.

  • GLM-5.2

    glm-5.2
    Not enough data
    30 days agoToday
    30-day uptime
    86.56%
    Median response duration (24h)
    7.3 s

  • DeepSeek V4 Flash 0731

    deepseek-v4-flash-0731
    Operational
    30 days agoToday
    30-day uptime
    96.11%
    Median response duration (24h)
    3.0 s

  • DeepSeek V4.1 Flash

    deepseek-v4.1-flash
    Operational
    30 days agoToday
    30-day uptime
    97.10%
    Median response duration (24h)
    6.1 s

  • Kimi K3

    kimi-k3
    Operational
    30 days agoToday
    30-day uptime
    97.72%
    Median response duration (24h)
    21.4 s

  • DeepSeek V4 Pro 0813

    deepseek-v4-pro-0813
    Operational
    30 days agoToday
    30-day uptime
    96.78%
    Median response duration (24h)
    9.4 s

  • Qwen 8 Max

    qwen3.8-max
    Not enough data
    30 days agoToday
    30-day uptime
    98.36%
    Median response duration (24h)
    2.3 s

  • GLM 5.3 Flash

    glm-5.3-flash
    Operational
    30 days agoToday
    30-day uptime
    99.10%
    Median response duration (24h)
    3.1 s

  • GLM 5.3

    glm-5.3
    Operational
    30 days agoToday
    30-day uptime
    99.71%
    Median response duration (24h)
    3.2 s

  • Gemini 3.8 Flash High

    gemini-3.8-flash-high
    Operational
    30 days agoToday
    30-day uptime
    94.06%
    Median response duration (24h)
    28.7 s

How we measure

Uptime is the share of successful requests versus requests that failed because of the service or its upstream. Requests rejected for an empty balance or quota, a model that is not allowed, or a malformed request do not count as downtime. Response duration is the total time until the response finishes, including streaming, so it depends on answer length. Status uses the last hour of data, or the last 24 hours when traffic is low.