Verified 2026-11-18 · rates follow the live catalog

Openference vs AZET — per-token rates compared

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.

The short answer

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.

AZET pricing

AZET plans as published
PlanPriceWhat it includes
Pay-as-you-goFrom $5Prepaid USD credit billed per token across 31 models. No subscription and no monthly minimum.

AZET and Openference side by side

AZET compared with Openference
AZETOpenference
Pricing modelPrepaid credit, billed per tokenPlan-based (included usage) + on-demand credits
Entry point$5 prepaid top-upFrom $20/mo (Minion)
Models included31 models: DeepSeek, GLM, Kimi, QwenOpen and frontier models
MiniMax modelsNoYes
Proprietary frontier modelsNoYes
Price relative to the model makerAt 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

Same model, per-token price

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).

Price per 1M tokens for models both AZET and Openference list
ModelAZET per 1M (in / out)Openference per 1M (in / out)AZET, one requestOpenference, one requestLower price
GLM-5.3$1.1319 / $3.5574$1.40 / $4.40$0.026$0.032AZET, 19% lower
GLM-5.2$1.1319 / $3.5574$1.40 / $4.40$0.026$0.032AZET, 19% lower
Kimi K3$2.4255 / $12.1275$3 / $15$0.061$0.075AZET, 19% lower
Kimi K2.7 Code$0.7681 / $3.234$0.95 / $4$0.019$0.023AZET, 19% lower
DeepSeek V4.1 Flash$0.2425 / $0.9702$0.30 / $1.20$0.0058$0.0072AZET, 19% lower
DeepSeek V4 Pro$1.0672 / $3.2017$1.32 / $3.96$0.025$0.03AZET, 19% lower
DeepSeek-V4-Flash-0731$0.3557 / $1.0672$0.44 / $1.32$0.0082$0.01AZET, 19% lower
Qwen3.8 27b$0.4042 / $2.4255$0.045 / $0.32$0.011$0.0012Openference, 88% lower
Qwen3.8 Flash$0.1213 / $0.38$0.075 / $0.235$0.0028$0.0017Openference, 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.

Questions

How much does AZET cost?

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.

Is AZET cheaper than Openference?

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.

Where do these figures come from?

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