Models
Routing targets
17 targets across 7 kinds; 4 answer through the OpenAI-compatible proxy and every one is routable for decisions and plans. Each LLM target shows the model it resolves to. Traffic is measured on this deployment over the last 7 days.
- targets
- 17
- executable
- 4
- requests · 7d
- 18
Recommended per use case
Live quotes of representative prompts against this catalogue, refreshed every minute. Quote your own prompt on the cost estimator.
- per request
- $0.00013
- quality
- 0.66
- latency
- 1.20 s
no rule matched · domain=0.50 action=0.50 complexity_fit=1.00 prior=0.70 · best_sim=0.06 top3_mean=0.02
- FastestMid-tier model$3.00/1M
- per request
- $0.00175
- quality
- 0.73
- latency
- 2.50 s
no rule matched · domain=1.00 action=1.00 complexity_fit=0.97 prior=0.93 · best_sim=0.04 top3_mean=0.03
- per request
- $0.00291
- quality
- 0.78
- latency
- 2.50 s
no rule matched · domain=1.00 action=1.00 complexity_fit=1.00 prior=0.93 · best_sim=0.19 top3_mean=0.10
- per request
- $0.0000553
- quality
- 0.76
- latency
- 1.20 s
no rule matched · domain=1.00 action=1.00 complexity_fit=1.00 prior=0.70 · best_sim=0.16 top3_mean=0.08
- per request
- $0.0000439
- quality
- 0.86
- latency
- 1.20 s
rule 'pii-stays-onprem' prefers · domain=0.50 action=1.00 complexity_fit=1.00 prior=0.70 · best_sim=0.19 top3_mean=0.09
- per request
- $0.0000678
- quality
- 0.89
- latency
- 1.20 s
rule 'pii-stays-onprem' prefers · domain=1.00 action=1.00 complexity_fit=1.00 prior=0.70 · best_sim=0.12 top3_mean=0.08
| Target | Kind | Model | Price / 1M | Context | Latency | Tokens 7d | Quality |
|---|---|---|---|---|---|---|---|
| On-prem private modelllm-onprem | llmexec | Llama 3.3 70B InstructMeta via azure/llm-onprem | $1.00 | - | 638.4 ms | 212new | 0.70 |
| UTranslation skillskill-translate | skill | - | $0.50 | - | 20.9 ms | 12new | 0.85 |
| UHuman escalation queuehuman-escalation | human | - | $500 | - | 15.9 ms | 1new | 0.90 |
| Frontier reasoning modelllm-frontier | llmexec | GPT-4.1OpenAI via azure/llm-frontier | $15 | 200K | 31.3 ms | 1new | 0.93 |
| UCoding agentcoding-agent | agent | - | $8.00 | - | 6000 ms | 0no traffic | 0.85 |
| Mid-tier modelllm-mid | llmexec | GPT-4.1 MiniOpenAI via azure/llm-mid | $3.00 | - | 900 ms | 0no traffic | 0.75 |
| Small fast modelllm-small | llmexec | GPT-4.1 NanoOpenAI via azure/llm-small | $0.20 | - | 300 ms | 0no traffic | 0.55 |
| UFriendly support personapersona-friendly-support | persona | - | free | - | 0 ms | 0no traffic | 0.75 |
| ULegal counsel personapersona-legal-counsel | persona | - | free | - | 0 ms | 0no traffic | 0.80 |
| USenior engineer personapersona-senior-engineer | persona | - | free | - | 0 ms | 0no traffic | 0.85 |
| UResearch agentresearch-agent | agent | - | $6.00 | - | 8000 ms | 0no traffic | 0.80 |
| USQL skillskill-sql | skill | - | $1.00 | - | 700 ms | 0no traffic | 0.80 |
| USummarization skillskill-summarize | skill | - | $0.80 | - | 1200 ms | 0no traffic | 0.82 |
| UCustomer support agentsupport-agent | agent | - | $2.00 | - | 1500 ms | 0no traffic | 0.80 |
| UCalculatortool-calculator | tool | - | free | - | 5 ms | 0no traffic | 0.99 |
| UImage generatortool-image-gen | tool | - | $40 | - | 9000 ms | 0no traffic | 0.85 |
| UCustomer notification workflowworkflow-notify-customer | workflow | - | free | - | 2000 ms | 0no traffic | 0.90 |
Add your own targets by editing targets.yaml; set metadata.model to a catalogue id such as anthropic/claude-sonnet-4 to link a target to its reference model. See the guide or the JSON behind this page.