modelstatus.dev

Qwen3 VL 235B A22B Instruct

qwen/qwen3-vl-235b-a22b-instruct

Compare with Qwen3 235B A22B Instruct 2507, Qwen3 30B A3B Instruct 2507, Qwen3 Coder 480B A35B

CompareQwen3 VL 235B A22B Instruct
  • Alibaba100%
    Healthy24h 99.90%
    p50
    644ms
    p99
    6.00s
    Throughput
    21 tps
    Context
    131K
    $/Mtok
    $0.260 / $1.04
    alibaba
  • Parasail99.36%
    Healthy24h 99.46%
    p50
    674ms
    p99
    5.38s
    Throughput
    14 tps
    Context
    131K
    $/Mtok
    $0.210 / $1.90
    parasail/fp8fp8
  • Venice96.59%
    Healthy24h 97.57%
    p50
    1.49s
    p99
    6.36s
    Throughput
    21 tps
    Context
    128K
    $/Mtok
    $0.210 / $1.90
    venice/fp8fp8
  • Novita90.62%
    Degraded24h 94.47%
    p50
    1.11s
    p99
    11.8s
    Throughput
    18 tps
    Context
    131K
    $/Mtok
    $0.300 / $1.50
    novita/bf16bf16
  • DeepInfra83.82%
    Down24h 96.60%
    p50
    921ms
    p99
    7.96s
    Throughput
    12 tps
    Context
    262K
    $/Mtok
    $0.200 / $0.880
    deepinfra/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 and p99 are OpenRouter's rolling 30-minute windows, averaged over the hour.

Throughput

Median tokens per second.