GLM 5.3: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100%◆ (best) | 99.97% | 98.33% | 98.33% | 1.05s | 9.70s | 73 tps | 1.0M | $1.40 / $4.40 |
| Healthy | 100%◆ (best) | 99.90% | 99.91% | 99.62% | 4.89s | 10.4s | 35 tps | 1.0M | $1.40 / $4.40 | |
| Healthy | 100%◆ (best) | 99.74% | 97.92% | 98.53% | 1.33s | 17.6s | 74 tps | 1.0M | $1.40 / $4.40 | |
| Healthy | 100%◆ (best) | 95.84% | 99.20% | 99.59% | 1.67s | 48.0s | 61 tps | 1.0M | $2.10 / $6.60 | |
| Healthy | 100%◆ (best) | 97.32% | 94.05% | 95.67% | 1.02s | 54.9s | 102 tps | 1.0M | $1.40 / $4.40 | |
| Degraded | 100%◆ (best) | 98.32% | 99.32% | 97.96% | 1.27s | 14.0s | 115 tps | 1.0M | $1.19 / $3.74 | |
DigitalOceandigitalocean | Healthy | 100%◆ (best) | 98.95% | 95.80% | 97.52% | 983ms | 5.90s | 46 tps | 1.0M | $1.08 / $3.39 |
Fireworksfireworks | Healthy | 100%◆ (best) | 99.59% | 99.73% | 99.56% | 1.40s | 38.4s | 46 tps | 1.0M | $1.40 / $4.40 |
Friendlifriendli | Healthy | 100%◆ (best) | 99.93% | 99.98% | 99.92% | 665ms | 6.97s | 111 tps | 1.0M | $1.26 / $3.96 |
| Healthy | 100%◆ (best) | 98.51% | 98.51% | 98.16% | 706ms | 9.23s | 69 tps | 1.0M | $1.07 / $4.09 | |
| Healthy | 100%◆ (best) | 99.53% | 99.63% | 98.98% | 1.83s | 7.55s | 65 tps | 262K | $0.910 / $3.08 | |
| Healthy | 100%◆ (best) | 95.92% | 96.82% | 94.22% | 1.37s | 42.6s | 63 tps | 980K | $1.05 / $4.20 | |
| Healthy | 100%◆ (best) | 99.38% | 98.63% | 98.63% | 3.10s | 37.4s | 20 tps | 1.0M | $0.952 / $2.99 | |
| Healthy | 100%◆ (best) | 99.21% | 99.03% | 97.09% | 1.08s | 94.3s | 84 tps | 1.0M | $1.40 / $4.40 | |
Phalaphala | Healthy | 100%◆ (best) | 99.71% | 99.02% | 98.95% | 1.29s | 15.4s | 79 tps | 1.0M | $1.05 / $3.30 |
| Healthy | 100%◆ (best) | 99.92% | 99.93% | 99.94% | 1.41s | 7.64s | 54 tps | 1.0M | $0.966 / $3.04 | |
| Healthy | 100%◆ (best) | 99.89% | 99.91% | 99.91% | 1.08s | 6.73s | 86 tps | 1.0M | $0.966 / $3.04 | |
| Healthy | 99.86% | 99.94% | 99.78% | 99.91% | 2.96s | 33.2s | 47 tps | 1.0M | $1.40 / $4.40 | |
| Healthy | 99.51% | 99.03% | 99.27% | 97.14% | 2.82s | 45.0s | 19 tps | 262K | $0.877 / $2.97 | |
| Healthy | 99.44% | 97.50% | 99.43% | 99.62% | 3.78s | 52.4s | 36 tps | 1.0M | $0.910 / $2.86 | |
Modalmodal | Healthy | 99.40% | 98.10% | 99.28% | 99.55% | 750ms | 33.3s | 73 tps | 1.0M | $1.40 / $4.40 |
| Healthy | 98.80% | 95.47% | 97.04% | 96.24% | 3.51s | 111s | 20 tps | 1.0M | $0.900 / $3.00 | |
Waferwafer | Healthy | 97.97% | 99.46% | 98.31% | 97.44% | 557ms | 17.2s | 113 tps | 1.0M | $1.40 / $4.40 |
Togethertogether | Healthy | 96.42% | 94.95% | 96.79% | 96.76% | 349ms◆ (best) | 52.9s | 120 tps◆ (best) | 1.0M | $1.40 / $4.40 |
| Degraded | 91.82% | 84.34% | 38.70% | 38.70% | 2.85s | 94.7s | 26 tps | 1.0M | $0.900 / $3.00 | |
| No data | — | 99.77% | 99.36% | 97.76% | 614ms | 2.23s | 105 tps | 1.0M | $1.30 / $4.40 | |
| No data | — | 100% | 100%◆ (best) | 100%◆ (best) | 624ms | 1.48s◆ (best) | 84 tps | 1.0M | $1.40 / $4.40 | |
Cloudflarecloudflare | No data | — | 99.95% | 99.78% | 99.72% | 4.18s | 69.0s | 38 tps | 1.3M | $1.40 / $4.40 |
| No data | — | — | — | 99.41% | — | — | — | 1.0M | $1.40 / $4.40 | |
| No data | — | — | — | 88.65% | — | — | — | 1.0M | $1.20 / $4.00 | |
| No data | — | — | — | 95.08% | — | — | — | 1.0M | $1.20 / $4.00 | |
Fireworksfireworks/us | No data | — | — | — | — | — | — | — | 1.0M | $2.10 / $6.60 |
| No data | — | 98.54% | 99.26% | 98.74% | 3.71s | 8.62s | 21 tps | 1.0M | $1.05 / $3.30 | |
| No data | — | — | 100%◆ (best) | 99.72% | — | — | — | 262K | $1.13 / $4.32 | |
| No data | — | 98.85% | 99.34% | 99.34% | 4.75s | 145s | 34 tps | 1.0M | $1.40 / $4.40 | |
| No data | — | — | 100%◆ (best) | 95.37% | — | — | — | 1.0M | $1.00 / $3.41 | |
| No data | — | 99.60% | 99.21% | 99.28% | 2.44s | 39.6s | 46 tps | 1.0M | $1.12 / $3.52 | |
StreamLakestreamlake | No data | — | — | — | 100%◆ (best) | — | — | — | 1.0M | $1.11 / $3.50 |
| No data | — | — | 99.65% | 98.85% | — | — | — | 1.0M | $0.700 / $2.20◆ (best) | |
Venicevenice | No data | — | 97.44% | 98.51% | 96.21% | 1.33s | 86.4s | 44 tps | 1.0M | $1.40 / $4.40 |
- Alibaba100%◆ (best)Healthy24h 99.97%
- 7d
- 98.33%
- 30d
- 98.33%
- p50
- 1.05s
- p99
- 9.70s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
alibaba - AtlasCloud100%◆ (best)Healthy24h 99.90%
- 7d
- 99.91%
- 30d
- 99.62%
- p50
- 4.89s
- p99
- 10.4s
- Throughput
- 35 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
atlas-cloud/fp8fp8 - BaseTen100%◆ (best)Healthy24h 99.74%
- 7d
- 97.92%
- 30d
- 98.53%
- p50
- 1.33s
- p99
- 17.6s
- Throughput
- 74 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
baseten/fp4fp4 - BaseTen100%◆ (best)Healthy24h 95.84%
- 7d
- 99.20%
- 30d
- 99.59%
- p50
- 1.67s
- p99
- 48.0s
- Throughput
- 61 tps
- Context
- 1.0M
- $/Mtok
- $2.10 / $6.60
baseten/fp8fp8 - Crusoe100%◆ (best)Healthy24h 97.32%
- 7d
- 94.05%
- 30d
- 95.67%
- p50
- 1.02s
- p99
- 54.9s
- Throughput
- 102 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
crusoe/fp4fp4 - Decart100%◆ (best)Degraded24h 98.32%
- 7d
- 99.32%
- 30d
- 97.96%
- p50
- 1.27s
- p99
- 14.0s
- Throughput
- 115 tps
- Context
- 1.0M
- $/Mtok
- $1.19 / $3.74
decart/fp4fp4 - DigitalOcean100%◆ (best)Healthy24h 98.95%
- 7d
- 95.80%
- 30d
- 97.52%
- p50
- 983ms
- p99
- 5.90s
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $1.08 / $3.39
digitalocean - Fireworks100%◆ (best)Healthy24h 99.59%
- 7d
- 99.73%
- 30d
- 99.56%
- p50
- 1.40s
- p99
- 38.4s
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
fireworks - Friendli100%◆ (best)Healthy24h 99.93%
- 7d
- 99.98%
- 30d
- 99.92%
- p50
- 665ms
- p99
- 6.97s
- Throughput
- 111 tps
- Context
- 1.0M
- $/Mtok
- $1.26 / $3.96
friendli - Inceptron100%◆ (best)Healthy24h 98.51%
- 7d
- 98.51%
- 30d
- 98.16%
- p50
- 706ms
- p99
- 9.23s
- Throughput
- 69 tps
- Context
- 1.0M
- $/Mtok
- $1.07 / $4.09
inceptron/fp4fp4 - Io Net100%◆ (best)Healthy24h 99.53%
- 7d
- 99.63%
- 30d
- 98.98%
- p50
- 1.83s
- p99
- 7.55s
- Throughput
- 65 tps
- Context
- 262K
- $/Mtok
- $0.910 / $3.08
io-net/fp8fp8 - Makora100%◆ (best)Healthy24h 95.92%
- 7d
- 96.82%
- 30d
- 94.22%
- p50
- 1.37s
- p99
- 42.6s
- Throughput
- 63 tps
- Context
- 980K
- $/Mtok
- $1.05 / $4.20
makora/fp4fp4 - Morph100%◆ (best)Healthy24h 99.38%
- 7d
- 98.63%
- 30d
- 98.63%
- p50
- 3.10s
- p99
- 37.4s
- Throughput
- 20 tps
- Context
- 1.0M
- $/Mtok
- $0.952 / $2.99
morph/fp8fp8 - Parasail100%◆ (best)Healthy24h 99.21%
- 7d
- 99.03%
- 30d
- 97.09%
- p50
- 1.08s
- p99
- 94.3s
- Throughput
- 84 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
parasail/fp8fp8 - Phala100%◆ (best)Healthy24h 99.71%
- 7d
- 99.02%
- 30d
- 98.95%
- p50
- 1.29s
- p99
- 15.4s
- Throughput
- 79 tps
- Context
- 1.0M
- $/Mtok
- $1.05 / $3.30
phala - Sail Research100%◆ (best)Healthy24h 99.92%
- 7d
- 99.93%
- 30d
- 99.94%
- p50
- 1.41s
- p99
- 7.64s
- Throughput
- 54 tps
- Context
- 1.0M
- $/Mtok
- $0.966 / $3.04
sail-research/fp8fp8 - Sail Research100%◆ (best)Healthy24h 99.89%
- 7d
- 99.91%
- 30d
- 99.91%
- p50
- 1.08s
- p99
- 6.73s
- Throughput
- 86 tps
- Context
- 1.0M
- $/Mtok
- $0.966 / $3.04
sail-research/usfp8 - Z.AI99.86%Healthy24h 99.94%
- 7d
- 99.78%
- 30d
- 99.91%
- p50
- 2.96s
- p99
- 33.2s
- Throughput
- 47 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
z-ai/fp8fp8 - Reka99.51%Healthy24h 99.03%
- 7d
- 99.27%
- 30d
- 97.14%
- p50
- 2.82s
- p99
- 45.0s
- Throughput
- 19 tps
- Context
- 262K
- $/Mtok
- $0.877 / $2.97
reka/fp8fp8 - Novita99.44%Healthy24h 97.50%
- 7d
- 99.43%
- 30d
- 99.62%
- p50
- 3.78s
- p99
- 52.4s
- Throughput
- 36 tps
- Context
- 1.0M
- $/Mtok
- $0.910 / $2.86
novita/fp8fp8 - Modal99.40%Healthy24h 98.10%
- 7d
- 99.28%
- 30d
- 99.55%
- p50
- 750ms
- p99
- 33.3s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
modal - DeepInfra98.80%Healthy24h 95.47%
- 7d
- 97.04%
- 30d
- 96.24%
- p50
- 3.51s
- p99
- 111s
- Throughput
- 20 tps
- Context
- 1.0M
- $/Mtok
- $0.900 / $3.00
deepinfra/fp4fp4 - Wafer97.97%Healthy24h 99.46%
- 7d
- 98.31%
- 30d
- 97.44%
- p50
- 557ms
- p99
- 17.2s
- Throughput
- 113 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
wafer - Together96.42%Healthy24h 94.95%
- 7d
- 96.79%
- 30d
- 96.76%
- p50
- 349ms◆ (best)
- p99
- 52.9s
- Throughput
- 120 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
together - InferenceNet91.82%Degraded24h 84.34%
- 7d
- 38.70%
- 30d
- 38.70%
- p50
- 2.85s
- p99
- 94.7s
- Throughput
- 26 tps
- Context
- 1.0M
- $/Mtok
- $0.900 / $3.00
inference-net/fp4fp4 - No data24h 99.77%
- 7d
- 99.36%
- 30d
- 97.76%
- p50
- 614ms
- p99
- 2.23s
- Throughput
- 105 tps
- Context
- 1.0M
- $/Mtok
- $1.30 / $4.40
akashml/fp8fp8 - No data24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 624ms
- p99
- 1.48s◆ (best)
- Throughput
- 84 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
baidu/fp8fp8 - No data24h 99.95%
- 7d
- 99.78%
- 30d
- 99.72%
- p50
- 4.18s
- p99
- 69.0s
- Throughput
- 38 tps
- Context
- 1.3M
- $/Mtok
- $1.40 / $4.40
cloudflare - No data24h —
- 7d
- —
- 30d
- 99.41%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
crusoe/fp8fp8 - No data24h —
- 7d
- —
- 30d
- 88.65%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $1.20 / $4.00
deepinfra/bf16bf16 - No data24h —
- 7d
- —
- 30d
- 95.08%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $1.20 / $4.00
deepinfra/fp8fp8 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $2.10 / $6.60
fireworks/us - No data24h 98.54%
- 7d
- 99.26%
- 30d
- 98.74%
- p50
- 3.71s
- p99
- 8.62s
- Throughput
- 21 tps
- Context
- 1.0M
- $/Mtok
- $1.05 / $3.30
gmicloud/fp8fp8 - No data24h —
- 7d
- 100%◆ (best)
- 30d
- 99.72%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $1.13 / $4.32
io-net/fp4fp4 - No data24h 98.85%
- 7d
- 99.34%
- 30d
- 99.34%
- p50
- 4.75s
- p99
- 145s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
mistral/nvfp4nvfp4 - No data24h —
- 7d
- 100%◆ (best)
- 30d
- 95.37%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $1.00 / $3.41
morph/fp4fp4 - No data24h 99.60%
- 7d
- 99.21%
- 30d
- 99.28%
- p50
- 2.44s
- p99
- 39.6s
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $1.12 / $3.52
siliconflow/fp8fp8 - No data24h —
- 7d
- —
- 30d
- 100%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $1.11 / $3.50
streamlake - No data24h —
- 7d
- 99.65%
- 30d
- 98.85%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $0.700 / $2.20◆ (best)
streamlake/fp8fp8 - No data24h 97.44%
- 7d
- 98.51%
- 30d
- 96.21%
- p50
- 1.33s
- p99
- 86.4s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
venice
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
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.