modelstatus.dev

DeepSeek V4 Flash Vision Exp

deepseek/deepseek-v4-flash-vision-expInputtextimageOutputtext

Compare its 7 providersIncident RSSCompare with DeepSeek V4 Flash 0731, DeepSeek V4 Flash 0423, DeepSeek V4.1 Flash

CompareDeepSeek V4 Flash Vision Exp
  • Healthy24h 99.72%
    7d
    99.68%
    30d
    95.97%
    p50
    1.09s
    p99
    5.47s
    Throughput
    105 tps
    Context
    1.0M
    $/Mtok
    $0.216 / $0.647
    deepinfra/fp8fp8
  • Healthy24h 99.99%
    7d
    99.88%
    30d
    99.40%
    p50
    2.95s
    p99
    9.97s
    Throughput
    117 tps
    Context
    1.0M
    $/Mtok
    $0.440 / $1.32
    gmicloud/fp8fp8
  • Novita100%
    Healthy24h 99.97%
    7d
    99.00%
    30d
    99.41%
    p50
    1.55s
    p99
    3.90s
    Throughput
    61 tps
    Context
    1.0M
    $/Mtok
    $0.440 / $1.32
    novita
  • Healthy24h 99.97%
    7d
    99.77%
    30d
    99.14%
    p50
    1.48s
    p99
    5.82s
    Throughput
    123 tps
    Context
    1.0M
    $/Mtok
    $0.440 / $1.32
    siliconflow/fp8fp8
  • No data24h —
    7d
    —
    30d
    93.35%
    p50
    —
    p99
    —
    Throughput
    —
    Context
    1.0M
    $/Mtok
    $0.440 / $1.32
    atlas-cloud/fp8fp8
  • No data24h —
    7d
    —
    30d
    99.95%
    p50
    —
    p99
    —
    Throughput
    —
    Context
    1.0M
    $/Mtok
    $0.440 / $1.32
    deepseek
  • No data24h —
    7d
    0.00%
    30d
    95.00%
    p50
    —
    p99
    —
    Throughput
    —
    Context
    1.0M
    $/Mtok
    $0.220 / $0.660
    fireworks

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, averaged over the hour.

Throughput

Median tokens per second.

Daily availability

One row per endpoint · one dot per day · last 90 days.

worst hour ≥ 95%90–95%< 90%no data

Colour by each endpoint's worst hour of each UTC day · 7 endpoints · 44 days with data, 29 endpoint-days below 90%.

Price changes