GLM 5.2: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 98.79% | 99.87% | 99.87% | 1.35s | 9.37s | 52 tps | 1.0M | $2.31 / $7.26 | |
| Healthy | 100%◆ (best) | 99.18% | 99.79% | 99.79% | 1.28s | 6.25s | 44 tps | 1.0M | $1.26 / $3.96 | |
| Healthy | 100%◆ (best) | 99.88% | 99.82% | 99.82% | 1.22s | 9.11s | 59 tps | 1.0M | $0.462 / $1.45◆ (best) | |
| Healthy | 100%◆ (best) | 99.40% | 99.39% | 99.39% | 2.08s | 39.1s | 66 tps | 1.0M | $2.10 / $6.60 | |
Cloudflarecloudflare | Healthy | 100%◆ (best) | 100% | 99.85% | 99.85% | 1.89s | 15.7s | 50 tps | 262K | $1.40 / $4.40 |
| Healthy | 100%◆ (best) | 99.68% | 99.76% | 99.76% | 750ms | 10.8s | 56 tps | 1.0M | $1.40 / $4.40 | |
Friendlifriendli | Healthy | 100%◆ (best) | 99.71% | 99.24% | 99.24% | 508ms◆ (best) | 7.05s | 85 tps | 1.0M | $1.40 / $4.40 |
| Healthy | 100%◆ (best) | 92.34% | 95.86% | 95.86% | 1.58s | 17.6s | 23 tps | 1.0M | $0.842 / $3.14 | |
| Healthy | 100%◆ (best) | 99.88% | 99.89% | 99.89% | 674ms | 11.5s | 71 tps | 262K | $1.40 / $4.40 | |
| Healthy | 100%◆ (best) | 96.08% | 99.71% | 99.71% | 1.76s | 11.5s | 37 tps | 1.0M | $1.26 / $3.00 | |
| Healthy | 100%◆ (best) | 99.59% | 99.64% | 99.64% | 4.18s | 43.1s | 7.0 tps | 1.0M | $0.500 / $3.15 | |
| Healthy | 100%◆ (best) | 99.12% | 99.71% | 99.71% | 1.41s | 18.8s | 34 tps | 1.0M | $1.19 / $3.74 | |
| Healthy | 100%◆ (best) | 98.27% | 98.57% | 98.57% | 1.89s | 18.5s | 45 tps | 1.0M | $1.40 / $4.40 | |
| Healthy | 100%◆ (best) | 99.87% | 99.63% | 99.63% | 1.07s | 12.0s | 52 tps | 1.0M | $1.40 / $4.40 | |
| Healthy | 99.95% | 95.12% | 99.47% | 99.47% | 1.20s | 7.53s | 44 tps | 1.0M | $0.742 / $2.33 | |
| Healthy | 99.93% | 99.50% | 99.29% | 99.29% | 1.69s | 28.8s | 29 tps | 1.0M | $0.750 / $2.40 | |
Fireworksfireworks | Healthy | 99.73% | 98.32% | 98.72% | 98.72% | 1.61s | 14.6s | 48 tps | 1.0M | $1.40 / $4.40 |
| Healthy | 99.64% | 99.18% | 99.70% | 99.70% | 2.22s | 13.3s | 40 tps | 1.0M | $0.476 / $1.50 | |
| Healthy | 99.47% | 98.67% | 99.59% | 99.59% | 823ms | 10.9s | 53 tps | 1.0M | $0.966 / $3.04 | |
DigitalOceandigitalocean | Healthy | 99.19% | 99.33% | 99.12% | 99.12% | 966ms | 16.8s | 44 tps | 262K | $0.700 / $2.20 |
| Healthy | 99.14% | 99.55% | 99.42% | 99.42% | 1.58s | 22.0s | 50 tps | 1.0M | $0.741 / $2.33 | |
| Healthy | 99.10% | 99.52% | 99.60% | 99.60% | 1.88s | 17.3s | 63 tps | 1.0M | $1.40 / $4.40 | |
Togethertogether | Healthy | 99.07% | 90.93% | 95.19% | 95.19% | 781ms | 18.2s | 50 tps | 512K | $1.40 / $4.40 |
| Healthy | 98.91% | 96.04% | 96.76% | 96.76% | 652ms | 13.9s | 76 tps | 1.0M | $0.750 / $2.90 | |
| Healthy | 98.14% | 97.57% | 98.45% | 98.45% | 1.18s | 12.2s | 56 tps | 1.0M | $0.684 / $2.28 | |
Fireworksfireworks/fast-us | Healthy | 98.08% | 95.88% | 96.77% | 96.77% | 2.00s | 16.3s | 94 tps | 1.0M | $2.10 / $6.60 |
| Healthy | 97.37% | 90.63% | 92.64% | 92.64% | 973ms | 18.2s | 42 tps | 203K | $0.600 / $2.00 | |
Fireworksfireworks/fast | Healthy | 97.22% | 96.20% | 97.47% | 97.47% | 1.35s | 8.90s | 86 tps | 1.0M | $2.10 / $6.60 |
| No data | — | — | 97.12% | 97.12% | — | — | — | 97K | $0.770 / $2.42 | |
Cloudflarecloudflare/fast | No data | — | 99.90% | 99.92%◆ (best) | 99.92%◆ (best) | 777ms | 3.62s◆ (best) | 127 tps◆ (best) | 262K | $2.10 / $6.60 |
| No data | — | 99.80% | 99.80% | 99.80% | 1.24s | 8.89s | 85 tps | 262K | $0.760 / $2.42 | |
| Degraded | — | 98.48% | 98.83% | 98.83% | 5.30s | 17.5s | 44 tps | 1.0M | $1.26 / $3.96 |
- Alibaba100%◆ (best)Healthy24h 98.79%
- 7d
- 99.87%
- 30d
- 99.87%
- p50
- 1.35s
- p99
- 9.37s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $2.31 / $7.26
alibaba/fastfp8 - AtlasCloud100%◆ (best)Healthy24h 99.18%
- 7d
- 99.79%
- 30d
- 99.79%
- p50
- 1.28s
- p99
- 6.25s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $1.26 / $3.96
atlas-cloud/fp8fp8 - Baidu100%◆ (best)Healthy24h 99.88%
- 7d
- 99.82%
- 30d
- 99.82%
- p50
- 1.22s
- p99
- 9.11s
- Throughput
- 59 tps
- Context
- 1.0M
- $/Mtok
- $0.462 / $1.45◆ (best)
baidu/fp8fp8 - BaseTen100%◆ (best)Healthy24h 99.40%
- 7d
- 99.39%
- 30d
- 99.39%
- p50
- 2.08s
- p99
- 39.1s
- Throughput
- 66 tps
- Context
- 1.0M
- $/Mtok
- $2.10 / $6.60
baseten/fastfp8 - Cloudflare100%◆ (best)Healthy24h 100%
- 7d
- 99.85%
- 30d
- 99.85%
- p50
- 1.89s
- p99
- 15.7s
- Throughput
- 50 tps
- Context
- 262K
- $/Mtok
- $1.40 / $4.40
cloudflare - Crusoe100%◆ (best)Healthy24h 99.68%
- 7d
- 99.76%
- 30d
- 99.76%
- p50
- 750ms
- p99
- 10.8s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
crusoe/fp8fp8 - Friendli100%◆ (best)Healthy24h 99.71%
- 7d
- 99.24%
- 30d
- 99.24%
- p50
- 508ms◆ (best)
- p99
- 7.05s
- Throughput
- 85 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
friendli - Morph100%◆ (best)Healthy24h 92.34%
- 7d
- 95.86%
- 30d
- 95.86%
- p50
- 1.58s
- p99
- 17.6s
- Throughput
- 23 tps
- Context
- 1.0M
- $/Mtok
- $0.842 / $3.14
morph/fp4fp4 - Parasail100%◆ (best)Healthy24h 99.88%
- 7d
- 99.89%
- 30d
- 99.89%
- p50
- 674ms
- p99
- 11.5s
- Throughput
- 71 tps
- Context
- 262K
- $/Mtok
- $1.40 / $4.40
parasail/fp4fp4 - Phala100%◆ (best)Healthy24h 96.08%
- 7d
- 99.71%
- 30d
- 99.71%
- p50
- 1.76s
- p99
- 11.5s
- Throughput
- 37 tps
- Context
- 1.0M
- $/Mtok
- $1.26 / $3.00
phala/fp8fp8 - Sail Research100%◆ (best)Healthy24h 99.59%
- 7d
- 99.64%
- 30d
- 99.64%
- p50
- 4.18s
- p99
- 43.1s
- Throughput
- 7.0 tps
- Context
- 1.0M
- $/Mtok
- $0.500 / $3.15
sail-research/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 99.12%
- 7d
- 99.71%
- 30d
- 99.71%
- p50
- 1.41s
- p99
- 18.8s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $1.19 / $3.74
siliconflow/fp8fp8 - Venice100%◆ (best)Healthy24h 98.27%
- 7d
- 98.57%
- 30d
- 98.57%
- p50
- 1.89s
- p99
- 18.5s
- Throughput
- 45 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
venice/fp8fp8 - Z.AI100%◆ (best)Healthy24h 99.87%
- 7d
- 99.63%
- 30d
- 99.63%
- p50
- 1.07s
- p99
- 12.0s
- Throughput
- 52 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
z-ai/fp8fp8 - GMICloud99.95%Healthy24h 95.12%
- 7d
- 99.47%
- 30d
- 99.47%
- p50
- 1.20s
- p99
- 7.53s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $0.742 / $2.33
gmicloud/fp8fp8 - DeepInfra99.93%Healthy24h 99.50%
- 7d
- 99.29%
- 30d
- 99.29%
- p50
- 1.69s
- p99
- 28.8s
- Throughput
- 29 tps
- Context
- 1.0M
- $/Mtok
- $0.750 / $2.40
deepinfra/fp4fp4 - Fireworks99.73%Healthy24h 98.32%
- 7d
- 98.72%
- 30d
- 98.72%
- p50
- 1.61s
- p99
- 14.6s
- Throughput
- 48 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
fireworks - StreamLake99.64%Healthy24h 99.18%
- 7d
- 99.70%
- 30d
- 99.70%
- p50
- 2.22s
- p99
- 13.3s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $0.476 / $1.50
streamlake/fp8fp8 - Alibaba99.47%Healthy24h 98.67%
- 7d
- 99.59%
- 30d
- 99.59%
- p50
- 823ms
- p99
- 10.9s
- Throughput
- 53 tps
- Context
- 1.0M
- $/Mtok
- $0.966 / $3.04
alibaba/fp8fp8 - DigitalOcean99.19%Healthy24h 99.33%
- 7d
- 99.12%
- 30d
- 99.12%
- p50
- 966ms
- p99
- 16.8s
- Throughput
- 44 tps
- Context
- 262K
- $/Mtok
- $0.700 / $2.20
digitalocean - Novita99.14%Healthy24h 99.55%
- 7d
- 99.42%
- 30d
- 99.42%
- p50
- 1.58s
- p99
- 22.0s
- Throughput
- 50 tps
- Context
- 1.0M
- $/Mtok
- $0.741 / $2.33
novita/fp8fp8 - BaseTen99.10%Healthy24h 99.52%
- 7d
- 99.60%
- 30d
- 99.60%
- p50
- 1.88s
- p99
- 17.3s
- Throughput
- 63 tps
- Context
- 1.0M
- $/Mtok
- $1.40 / $4.40
baseten/fp8fp8 - Together99.07%Healthy24h 90.93%
- 7d
- 95.19%
- 30d
- 95.19%
- p50
- 781ms
- p99
- 18.2s
- Throughput
- 50 tps
- Context
- 512K
- $/Mtok
- $1.40 / $4.40
together - Inceptron98.91%Healthy24h 96.04%
- 7d
- 96.76%
- 30d
- 96.76%
- p50
- 652ms
- p99
- 13.9s
- Throughput
- 76 tps
- Context
- 1.0M
- $/Mtok
- $0.750 / $2.90
inceptron/fp4fp4 - Decart98.14%Healthy24h 97.57%
- 7d
- 98.45%
- 30d
- 98.45%
- p50
- 1.18s
- p99
- 12.2s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $0.684 / $2.28
decart/fp4fp4 - Fireworks98.08%Healthy24h 95.88%
- 7d
- 96.77%
- 30d
- 96.77%
- p50
- 2.00s
- p99
- 16.3s
- Throughput
- 94 tps
- Context
- 1.0M
- $/Mtok
- $2.10 / $6.60
fireworks/fast-us - Ambient97.37%Healthy24h 90.63%
- 7d
- 92.64%
- 30d
- 92.64%
- p50
- 973ms
- p99
- 18.2s
- Throughput
- 42 tps
- Context
- 203K
- $/Mtok
- $0.600 / $2.00
ambient/fp8fp8 - Fireworks97.22%Healthy24h 96.20%
- 7d
- 97.47%
- 30d
- 97.47%
- p50
- 1.35s
- p99
- 8.90s
- Throughput
- 86 tps
- Context
- 1.0M
- $/Mtok
- $2.10 / $6.60
fireworks/fast - No data24h —
- 7d
- 97.12%
- 30d
- 97.12%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 97K
- $/Mtok
- $0.770 / $2.42
akashml/fp8fp8 - No data24h 99.90%
- 7d
- 99.92%◆ (best)
- 30d
- 99.92%◆ (best)
- p50
- 777ms
- p99
- 3.62s◆ (best)
- Throughput
- 127 tps◆ (best)
- Context
- 262K
- $/Mtok
- $2.10 / $6.60
cloudflare/fast - No data24h 99.80%
- 7d
- 99.80%
- 30d
- 99.80%
- p50
- 1.24s
- p99
- 8.89s
- Throughput
- 85 tps
- Context
- 262K
- $/Mtok
- $0.760 / $2.42
coreweave/fp4fp4 - Degraded24h 98.48%
- 7d
- 98.83%
- 30d
- 98.83%
- p50
- 5.30s
- p99
- 17.5s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $1.26 / $3.96
waferfp4
Measured 2m 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.