modelstatus.dev

Qwen3.8 27B

qwen/qwen3.8-27b

Compare with Qwen3.8 2.4T A95B, Qwen-Plus, Qwen3.8 Max

CompareQwen3.8 27B
  • Healthy24h 99.86%
    7d
    99.88%
    30d
    99.88%
    p50
    1.69s
    p99
    84.4s
    Throughput
    51 tps
    Context
    1.0M
    $/Mtok
    $0.575 / $3.45
    alibaba
  • Io Net100%
    Healthy24h 98.57%
    7d
    99.12%
    30d
    99.12%
    p50
    2.27s
    p99
    12.0s
    Throughput
    21 tps
    Context
    66K
    $/Mtok
    $0.480 / $3.40
    io-net/fp8fp8
  • Healthy24h 98.91%
    7d
    98.48%
    30d
    98.48%
    p50
    885ms
    p99
    22.9s
    Throughput
    23 tps
    Context
    262K
    $/Mtok
    $0.450 / $3.20
    parasail/fp8fp8
  • Reka100%
    Healthy24h 99.70%
    7d
    99.48%
    30d
    99.48%
    p50
    1.10s
    p99
    7.78s
    Throughput
    57 tps
    Context
    262K
    $/Mtok
    $0.450 / $3.20
    reka/fp8fp8
  • Venice100%
    Healthy24h 99.75%
    7d
    99.21%
    30d
    99.21%
    p50
    976ms
    p99
    31.3s
    Throughput
    93 tps
    Context
    262K
    $/Mtok
    $0.450 / $3.20
    venice/fp8fp8
  • Chutes96.49%
    Degraded24h 97.84%
    7d
    97.76%
    30d
    97.76%
    p50
    4.33s
    p99
    49.1s
    Throughput
    14 tps
    Context
    262K
    $/Mtok
    $0.400 / $3.00
    chutes/fp8fp8
  • AkashML93.70%
    Degraded24h 97.88%
    7d
    98.63%
    30d
    98.63%
    p50
    4.56s
    p99
    34.7s
    Throughput
    24 tps
    Context
    262K
    $/Mtok
    $0.450 / $3.20
    akashml/bf16bf16

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

Price changes

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.