Loading comparison…
HEAD TO HEAD
Featherless AI: simple-jev and Standard Thinking: standard-one, compared on what they decide, where they run, what they cost and what their publishers report.
| Property | featherless-ai/simple-jev | standard-thinking/standard-one |
|---|---|---|
| Summary | Featherless AI's open server that turns a stock Hugging Face language model into a typed-decision endpoint. It reads the next-token logits for each question and builds the Choice, Score or Noul answer itself; no classifier head is trained and no JSON is generated. | Standard Thinking's decision model. Ministral 3 8B with a merged LoRA, served through stock SGLang and an open jev-adapter that exposes /v1/systemone and scores the supplied options in one forward pass with per-type temperatures. |
| Decides | choice, score, noul, classify, route | choice, score, noul, classify, route |
| Architecture | simple-jev | standard-one |
| Fine-tuned from | google/gemma-4-26b-a4b-it | mistralai/ministral-3-8b-instruct-2512-bf16 |
| License | apache-2.0 | apache-2.0 |
| Availability | Open weights + hosted API | Open weights + hosted API |
| Hosted by | Featherless AI | Standard Thinking |
| Input price | — | — |
| Decision accuracy | — | 71.1% |
| Calibration error | — | — |
| Valid action rate | — | — |
| Median latency | — | 25.8 ms |
| p95 latency | — |
Figures are from each model’s manifest; accuracy and latency are what the publishers report, on their own suites and hardware. Add a third model.
simple-jev is from Featherless AI and standard-one from Standard Thinking. Both have open weights and a hosted API. Both answer choice, score, noul, classify and route questions.
Only standard-one publishes an accuracy figure (71.1% on typed-decisions suite, 400 cases / 2,000 decisions (maker's served run, v2)); simple-jev does not, so there is no comparison to make without your own test.
simple-jev: Hosted, price not published, or free to self-host. standard-one: Hosted, price not published, or free to self-host. Open weights cost nothing per call beyond your own hardware.
Yes, both: systemone pull featherless-ai/simple-jev and systemone pull standard-thinking/standard-one download the weights.
| 41.9 ms |
| Evaluation suite | — | typed-decisions suite, 400 cases / 2,000 decisions (maker's served run, v2) |
|---|
| Latest version | 2026.09 | 2.0.0 |
|---|
| Variants | — | LICENSE, NOTICE, SHA256SUMS, evidence, server |
|---|
| Size of latest version | — | 16.6 GB |
|---|
| Files | 0 | 71 |
|---|
| Downloads | 0 | 0 |
|---|
| Stars | 0 | 0 |
|---|
| Tags | system-one, training-free, logit-readout, gemma, qwen | system-one, mistral, lora, sglang, multilingual, jevbench, 8b |
|---|
| Updated | Sep 30, 2026 | Sep 30, 2026 |
|---|