Verified 2026-11-18 · rates follow the live catalog
AZET is a prepaid, per-token API for 31 models — DeepSeek, GLM, Kimi and Qwen, plus a small model of its own — behind one OpenAI-compatible key. It says it charges its cost plus about 5% and never more than the model maker's own price; most listed models sit 19% under the maker's rate. There is no subscription and top-ups start at $5. MiniMax models are not in its catalog. Openference adds plans with bundled usage and team seats; compare the per-token rates below for the models both list.
Hosted gateway · Pay-as-you-go (prepaid from $5) · Checked 19 November 2026 against AZET Models and Pricing.
Pick AZET for prepaid pay-per-token GLM, DeepSeek, Kimi and Qwen. Pick Openference for a monthly plan with bundled usage, MiniMax and frontier models, and team seats.
| Plan | Price | What it includes |
|---|---|---|
| Pay-as-you-go | From $5 | Prepaid USD credit billed per token across 31 models. No subscription and no monthly minimum. |
| AZET | Openference | |
|---|---|---|
| Pricing model | Prepaid credit, billed per token | Plan-based (included usage) + on-demand credits |
| Entry point | $5 prepaid top-up | From $20/mo (Minion) |
| Models included | 31 models: DeepSeek, GLM, Kimi, Qwen | Open and frontier models |
| MiniMax models | No | Yes |
| Proprietary frontier models | No | Yes |
| Price relative to the model maker | At or below the maker's own price (about 19% under on most models) | No per-request fee and no markup on listed rates; 5% processing fee on credit top-ups |
On the 9 models both list, AZET has the lower price for a typical request on 7 and Openference on 2. The widest gap in Openference's favour is Qwen3.8 27b (88% lower); the widest in AZET's favour is GLM-5.3 (19% lower).
| Model | AZET per 1M (in / out) | Openference per 1M (in / out) | AZET, one request | Openference, one request | Lower price |
|---|---|---|---|---|---|
| GLM-5.3 | $1.1319 / $3.5574 | $1.40 / $4.40 | $0.026 | $0.032 | AZET, 19% lower |
| GLM-5.2 | $1.1319 / $3.5574 | $1.40 / $4.40 | $0.026 | $0.032 | AZET, 19% lower |
| Kimi K3 | $2.4255 / $12.1275 | $3 / $15 | $0.061 | $0.075 | AZET, 19% lower |
| Kimi K2.7 Code | $0.7681 / $3.234 | $0.95 / $4 | $0.019 | $0.023 | AZET, 19% lower |
| DeepSeek V4.1 Flash | $0.2425 / $0.9702 | $0.30 / $1.20 | $0.0058 | $0.0072 | AZET, 19% lower |
| DeepSeek V4 Pro | $1.0672 / $3.2017 | $1.32 / $3.96 | $0.025 | $0.03 | AZET, 19% lower |
| DeepSeek-V4-Flash-0731 | $0.3557 / $1.0672 | $0.44 / $1.32 | $0.0082 | $0.01 | AZET, 19% lower |
| Qwen3.8 27b | $0.4042 / $2.4255 | $0.045 / $0.32 | $0.011 | $0.0012 | Openference, 88% lower |
| Qwen3.8 Flash | $0.1213 / $0.38 | $0.075 / $0.235 | $0.0028 | $0.0017 | Openference, 38% lower |
A request here is a 20,000-token prompt answered with 1,000 tokens, with no prompt caching. AZET's rates are as published on AZET Models and Pricing, read 19 November 2026. Openference's are read from the live catalog at the highest rate of the day, so a model with a scheduled off-peak discount costs less at other hours than shown. Prices on both sides change; check the sources before you commit.
Pay-as-you-go (prepaid from $5). The plan table above lists each tier as AZET publishes it, read from AZET Models and Pricing on 19 November 2026.
On the 9 models both list, AZET has the lower price for a typical request on 7 and Openference on 2. The table above shows each model; which side is cheaper overall depends on which models you actually run.
AZET's from its own pages on 19 November 2026; Openference's from its live catalog and public plans. Both change, so confirm on the linked page before you commit.
More on the comparison index, the current plans and the model catalog.
Model catalog · Pricing · Comparisons · Cost guides · Integration guides · FAQ
Guides: GLM coding plan alternative · Free AI API · LLM gateway · LLM proxy · Cheap AI API · OpenAI-compatible API
Model APIs: GLM API · DeepSeek API · Kimi API · Qwen API · MiniMax API