Insufficient evidence to determine model identity. API, speed, tool, and token observations are recorded separately and do not establish identity.
All primary checks passed in this run.
The basic API call succeeded and returned a recognizable response.
No issues were recorded in the checks performed. Untested capabilities are outside this conclusion.
API results and model identity are assessed separately. Insufficient evidence does not mean a failed request or a fake model.
No reference samples under matching conditions are available for this model.
Model self-identification and individual answers do not establish authenticity. Official references only reflect API behavior at the time of sampling.
Rule version：2026-09-14.v1Reported usage is used; estimates appear only when none is reported.
Several runs in a row to see whether repeated content is billed at the cheaper cache rate; 0% means no cache discount this time.
Caching was used, but only a little; check your bill to see how much it really saves.
We test how much it can read at once, then compare that with the officially stated length.
All tested context tiers (32K) returned reliably.
Expand any row for details; anything private has been removed.
minimax/minimax-m3 is visible in the API model list.
This test ran 2 times; result: Passed.
The basic API call succeeded and returned a recognizable response.
This test ran 1 times; result: Passed.
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 222 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 181 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 193 tokens of which 128 tokens were billed at the cache price
按官方接口的标准调用，看返回是否规范。 1 次都对。
This test ran 1 times: 1 passed. This test used 231 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 204 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 249 tokens of which 128 tokens were billed at the cache price
核对它报回来的用量和缓存计费。 1 次都对。
This test ran 1 times: 1 passed. This test used 194 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 260 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 259 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 237 tokens of which 128 tokens were billed at the cache price
针对这个模型再做一项专门检测。 5 次都对。
This test ran 5 times: 5 passed. This test used 14.2K tokens of which 8.2K tokens were billed at the cache price
32K 长度的内容能稳定读完，这就是本次上下文实测的依据。
This test ran 1 times: 1 passed. Content length used: 32K This test used 16.9K tokens of which 128 tokens were billed at the cache price
还没有人评论，来写第一条。
Before you buy AI, check MozhenzhenInsufficient evidence to determine model identity. API, speed, tool, and token observations are recorded separately and do not establish identity.
The basic API call succeeded and returned a recognizable response.
No issues were recorded in the checks performed. Untested capabilities are outside this conclusion.
API results and model identity are assessed separately. Insufficient evidence does not mean a failed request or a fake model.
No reference samples under matching conditions are available for this model.
Model self-identification and individual answers do not establish authenticity. Official references only reflect API behavior at the time of sampling.
Rule version：2026-09-14.v1Reported usage is used; estimates appear only when none is reported.
Several runs in a row to see whether repeated content is billed at the cheaper cache rate; 0% means no cache discount this time.
Caching was used, but only a little; check your bill to see how much it really saves.
We test how much it can read at once, then compare that with the officially stated length.
All tested context tiers (32K) returned reliably.
Expand any row for details; anything private has been removed
minimax/minimax-m3 is visible in the API model list.
This test ran 2 times; result: Passed.
The basic API call succeeded and returned a recognizable response.
This test ran 1 times; result: Passed.
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 222 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 181 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 193 tokens of which 128 tokens were billed at the cache price
按官方接口的标准调用，看返回是否规范。 1 次都对。
This test ran 1 times: 1 passed. This test used 231 tokens of which 128 tokens were billed at the cache price
用几道题看它的回答水平像不像官方模型。 1 次都对。
This test ran 1 times: 1 passed. This test used 204 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 249 tokens of which 128 tokens were billed at the cache price
核对它报回来的用量和缓存计费。 1 次都对。
This test ran 1 times: 1 passed. This test used 194 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 260 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 259 tokens of which 128 tokens were billed at the cache price
Asked the model to identify itself; all 1 responses matched the expected name.
This test ran 1 times: 1 passed. This test used 237 tokens of which 128 tokens were billed at the cache price
针对这个模型再做一项专门检测。 5 次都对。
This test ran 5 times: 5 passed. This test used 14.2K tokens of which 8.2K tokens were billed at the cache price
32K 长度的内容能稳定读完，这就是本次上下文实测的依据。
This test ran 1 times: 1 passed. Content length used: 32K This test used 16.9K tokens of which 128 tokens were billed at the cache price
还没有人评论，来写第一条。
发表评论