LLM gateways & inference providers — compare pricing, reliability & setup
Generated 2026-11-03 12:14 UTC · JSON
For unified access (one api, many models), the current pick is DeepInfra (score 79.75). See cost by profile below.
Ranking — Unified access (one API, many models)
| Vendor | Score | gpt-oss-120B (10M in / 2M out) | Free | Agent-friction | Verified |
|---|---|---|---|---|---|
| DeepInfra | 79.75 | $0.71 | yes | 65 | 2026-11-03 |
| Fireworks AI | 74.29 | $2.70 | yes | 75 | 2026-11-03 |
| Together AI | 74.29 | $2.70 | yes | 75 | 2026-11-03 |
| OpenRouter | 72.79 | $2.70 | yes | 65 | 2026-11-03 |
| Groq | 55.29 | $2.70 | yes | 65 | 2026-11-03 |
| Cerebras | 47.25 | $5 | yes | 65 | 2026-11-03 |
Agent-friction — can a service be set up without a human?
Each check and its weight is published; the column total is the 0–100 agent-friction score.
| Check | DeepInfra | Fireworks AI | Together AI | OpenRouter | Groq | Cerebras |
|---|---|---|---|---|---|---|
| Test credentials without a credit card (20) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Sandbox or test mode (15) | × | × | × | × | × | × |
| Setup with no human-only steps (15) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Official CLI (10) | × | ✓ | ✓ | × | × | × |
| Official MCP server (10) | × | × | × | × | × | × |
| Markdown docs, current with the SDK (10) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Typed SDK for main languages (10) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Error messages that name the fix (10) | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Score | 65 | 75 | 75 | 65 | 65 | 65 |
How this category is scored
Per use case, weights published in the data: fit 35% · cost 15% · integration 15% · agent_friction 15% · health 12% · lock_in 8%. Affiliate status is never an input.