modelstatus.dev

GLM 4.6

z-ai/glm-4.6

Compare with GLM 4.6V, GLM 4.5 Air, GLM 4.5V

CompareGLM 4.6
  • Healthy24h 99.92%
    7d
    99.17%
    30d
    99.17%
    p50
    488ms
    p99
    8.81s
    Throughput
    50 tps
    Context
    203K
    $/Mtok
    $0.500 / $2.00
    deepinfra/fp4fp4
  • Venice100%
    Healthy24h 99.31%
    7d
    98.76%
    30d
    98.76%
    p50
    1.05s
    p99
    11.5s
    Throughput
    32 tps
    Context
    198K
    $/Mtok
    $0.430 / $1.75
    venice/fp4fp4
  • Z.AI97.50%
    Degraded24h 95.53%
    7d
    91.77%
    30d
    91.77%
    p50
    1.06s
    p99
    25.3s
    Throughput
    21 tps
    Context
    203K
    $/Mtok
    $0.600 / $2.20
    z-ai/fp4fp4
  • Novita91.11%
    Degraded24h 97.23%
    7d
    96.74%
    30d
    96.74%
    p50
    1.32s
    p99
    37.5s
    Throughput
    27 tps
    Context
    205K
    $/Mtok
    $0.550 / $2.20
    novita/bf16bf16
  • Degraded24h 92.88%
    7d
    97.11%
    30d
    97.11%
    p50
    660ms
    p99
    20.8s
    Throughput
    44 tps
    Context
    203K
    $/Mtok
    $0.600 / $2.20
    atlas-cloud/fp8fp8

Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.

Availability

Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.

Time to first token

Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window.

Throughput

Median tokens per second.