System status. Live telemetry.
Whether each model behind phantom-1 is answering now, from provider uptime and latency telemetry the upstream reports.
- Upstream uptime probed continuously
- Multi-provider automatic failover
- Zero-storage cryptographic receipts
Model status right now
Probe windows: 15 minutes and 24 hours
Routing Tier Models
- All 10 models are up.
| Runs | Status | Uptime | Providers |
|---|---|---|---|
Chat, economy model nvidia/nemotron-3.5-lightning-free | Up | 99.97% | 1 |
Chat, standard model deepseek/deepseek-v4.1-flash | Up | 100% | 22 |
Chat, frontier model anthropic/claude-sonnet-5.5 | Up | 100% | 4 |
Chat with an image alibaba/qwen3.7-flash | Up | 100% | 1 |
Embeddings openai/text-embedding-3-small | Up | 100% | 2 |
Rerank voyage/rerank-3 | Up | 100% | 1 |
Images bfl/flux-pro-1.1 | Up | 100% | 1 |
Video alibaba/wan-v2.6-t2v | Up | n/a | 1 |
Speech openai/tts-1 | Up | 100% | 1 |
Transcription openai/whisper-1 | Up | 100% | 1 |
Uptime is the best provider’s share of requests that worked in the last 15 minutes, or the last day when a provider doesn’t report 15 minutes. Up means at least 95%, degraded at least 50%, down less than that or no provider. A model whose providers report no uptime counts as up.
Providers by Model
Each provider the upstream lists for a model, best uptime first. Calls try the cheapest working provider first unless you set sort.
Chat, economy model
Upnvidia/nemotron-3.5-lightning-free
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| nvidia | 99.97% | 99.94% | 418 ms | Free | Free |
Chat, standard model
Updeepseek/deepseek-v4.1-flash
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| alibaba | 100% | 99.94% | 2,637 ms | $0.16 | $0.63 |
| boundless | 100% | 100% | 546 ms | $0.21 | $1.05 |
| deepinfra | 100% | 99.95% | 888 ms | $0.15 | $0.44 |
| digitalocean | 100% | 99.99% | 639 ms | $0.32 | $1.26 |
| fireworks | 100% | 99.99% | 895 ms | $0.32 | $1.26 |
| gmicloud | 100% | 100% | 1,174 ms | $0.19 | $0.76 |
| inference-net | 100% | 100% | 368 ms | $0.32 | $1.05 |
| modal | 100% | 100% | 850 ms | $0.32 | $1.26 |
| morph | 100% | 99.99% | 1,399 ms | $0.053 | $0.20 |
| novita | 100% | 100% | 1,003 ms | $0.20 | $0.82 |
| parasail | 100% | 100% | 288 ms | $0.32 | $1.26 |
| pareto | 100% | 99.99% | 1,504 ms | $0.079 | $0.32 |
| particle | 100% | 100% | 993 ms | $0.074 | $1.05 |
| relace | 100% | 99.98% | 894 ms | $0.037 | $0.70 |
| runinfra | 100% | 100% | 484 ms | $0.12 | $0.49 |
| runware | 100% | 99.95% | 1,169 ms | $0.16 | $0.63 |
| tencent | 100% | 99.93% | 1,678 ms | $0.16 | $0.63 |
| togetherai | 100% | 99.99% | 430 ms | $0.32 | $1.26 |
| vertex | 100% | 99.95% | 3,812 ms | $0.32 | $1.26 |
| wafer | 100% | 99.98% | 834 ms | $0.20 | $0.68 |
| baseten | 99.98% | 99.93% | 481 ms | $0.32 | $1.26 |
| deepseek | 99.96% | 99.97% | 1,800 ms | $0.16 | $0.63 |
Chat, frontier model
Upanthropic/claude-sonnet-5.5
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| anthropic | 100% | 99.99% | 1,139 ms | $2.10 | $10.50 |
| bedrock | 100% | 99.79% | 1,613 ms | $2.10 | $10.50 |
| claudeaws | 100% | 99.99% | 1,036 ms | $2.10 | $10.50 |
| vertexAnthropic | 100% | 100% | 530 ms | $2.10 | $10.50 |
Chat with an image
Upalibaba/qwen3.7-flash
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| alibaba | 100% | 99.99% | 2,330 ms | $0.032 | $0.14 |
Embeddings
Upopenai/text-embedding-3-small
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M |
|---|---|---|---|---|
| azure | 100% | 99.99% | n/a | $0.021 |
| openai | 100% | 99.99% | n/a | $0.021 |
Rerank
Upvoyage/rerank-3
| Provider | Uptime 15m | Uptime 1d | First token | Input / 1M |
|---|---|---|---|---|
| voyage | 100% | 100% | n/a | $0.053 |
Images
Upbfl/flux-pro-1.1
| Provider | Uptime 15m | Uptime 1d | First token |
|---|---|---|---|
| bfl | n/a | 100% | n/a |
Video
Upalibaba/wan-v2.6-t2v
| Provider | Uptime 15m | Uptime 1d | First token |
|---|---|---|---|
| alibaba | n/a | n/a | n/a |
Speech
Upopenai/tts-1
| Provider | Uptime 15m | Uptime 1d | First token |
|---|---|---|---|
| openai | 100% | 100% | n/a |
Transcription
Upopenai/whisper-1
| Provider | Uptime 15m | Uptime 1d | First token |
|---|---|---|---|
| openai | 100% | 100% | n/a |
First token is the median time to the first token over the last hour. Prices include Phantom’s markup. Image, video and audio models bill per unit. n/a means the upstream reported no figure.
Check from code
GET /v1/models/status returns this reading for any model id, without a key. For phantom-1, add job to pick which of its models to check.
curl "https://phantom.codes/v1/models/status?model=anthropic/claude-sonnet-5.5" curl "https://phantom.codes/v1/models/status?model=phantom-1&job=embedding"