Model Selection Guide
As of 2026-08-25, the Brivionix marketplace lists 9 models. All are pay-as-you-go and use the openai /v1/chat/completions endpoint. Capability labels below follow the current pricing page; the Grok cards currently show no capability labels.
Quick Recommendation
| Use case | Recommended model | Why |
|---|---|---|
| Lowest cost with full marketplace labels | gpt-5.6-luna | $1.25 input / $10 output; tagged 1.1m / 400k, Files, Reasoning, Tools, Vision |
| Balanced price and labeled capabilities | gpt-5.6-terra | $3 input / $24 output with the full label set |
| Highest-priced full-label option | gpt-5.6-sol | $6 input / $48 output with the full label set |
| Coding tasks | gpt-5.3-codex | Tagged Reasoning, Tools, and Vision; explicitly code-oriented model name |
| Long-context general work | gpt-5.5 | The page labels it 1.1M, the largest single context label shown |
| Grok workflows | grok-4.6 | Current Grok model on the pricing page |
Model Comparison
| Model | Context / Marketplace Tags | Capabilities | Input (USD/1M) | Output (USD/1M) | Cache Read | Cache Create |
|---|---|---|---|---|---|---|
gpt-5.3-codex | 400K | Reasoning, Tools, Vision | $2.0590 | $16.4700 | $0.2060 | — |
gpt-5.4 | 400K | Reasoning, Tools, Files, Vision | $2.9410 | $17.6470 | $0.2940 | — |
gpt-5.4-mini | 400K | reasoning, tools, files, vision | $2.3410 | $14.0470 | $0.2140 | — |
gpt-5.5 | 1.1M | Reasoning, Tools, Files, Vision | $5.8820 | $35.2940 | $0.5880 | — |
gpt-5.6-luna | 1.1M / 400K | 1.1m, 400k, Files, Reasoning, Tools, Vision | $1.2500 | $10.0000 | $0.2000 | $2.0000 |
gpt-5.6-sol | 1.1M / 400K | 1.1m, 400k, Files, Reasoning, Tools, Vision | $6.0000 | $48.0000 | $0.6500 | $7.2500 |
gpt-5.6-terra | 1.1M / 400K | 1.1m, 400k, Files, Reasoning, Tools, Vision | $3.0000 | $24.0000 | $0.3000 | $3.2500 |
grok-4.5 | — | — | $4.0000 | $12.0000 | $1.0000 | — |
grok-4.6 | — | — | $4.0000 | $12.0000 | $1.0000 | — |
Prices are USD / 1M Tokens. The 1.1m and 400k combination is recorded exactly as shown on the pricing page; it is not interpreted as a single context limit. Verify effective limits from actual responses when needed.
Groups
This documentation uses the default group. When creating a token, choose default. Access also depends on the token's model whitelist and an exact model-name match with the pricing page.
Selection Tips
- Start with
gpt-5.6-lunawhen cost is the priority. - Choose
gpt-5.6-terrafor a balance of price and the full marketplace label set. - Choose
gpt-5.3-codexfor code generation, code review, and programming workflows. - For long context, choose
gpt-5.5(1.1M) or one of the 5.6 models carrying the 1.1m label. - For Grok CLI or Grok-specific workflows, choose
grok-4.6; its card currently has no capability labels, so verify it in Playground first.
Cache Prices
All 9 models currently show a cache-read price. gpt-5.6-luna, gpt-5.6-sol, and gpt-5.6-terra also show cache-create prices. For production billing, use the pricing page and Usage Logs as the source of truth.