Updated
jevainews.com
Jev News
News, benchmarks, and API prices for TypeSafe's System One model
Jev is TypeSafe's hosted System One model. This desk records news, benchmarks, and listed API prices for that call, and for other hosted APIs and open checkpoints that return a choice, a score, or a yes/no. The catalog is decision models. What Jev is is the page for that call. News is the archive. Timeline groups the same stories by date. Compare keeps each test on its own row. This site is not affiliated with TypeSafe.
returns
Confidence on Choice and Score comes from the returned probabilities. Noul returns a probability between 0 and 1, without a separate confidence. Written out in What Jev returns.
Hosted calls
A short list. Venice, Liquid d1, Drex, GLiDE, Tev1, and Mercury Decide are on the full catalog, with the figure each page printed.
| Call | Who | One figure |
|---|---|---|
| jev-1.13.0 | TypeSafe, POST /v1/systemone | $0.042 per million input tokens, output free. Date headers 08:49:11 GMT to 08:49:40 GMT October 26, jev-latest named jev-1.14.0. On October 16, 2026, at 00:28 UTC, jev-latest still returned this id. Pricing. |
| jev-1.14.0 | TypeSafe, jev-latest and jev-preview | Opened October 16, at 08:19 UTC, one Box for every vehicle returned one rectangle, n = 3. TypeLLM's cookbook prints six vehicle boxes, including Tesla's flying Roadster. Jev Day 2026. |
| Decisions API | OpenAI, POST /v1/decisions | Generally available October 22. Date headers 02:23:12 GMT and 02:23:13 GMT October 23, two POSTs returned HTTP 200. Guide $0.10 uncached, $0.01 cached, output uncharged. OpenAI Jev. |
| Clef, Clef-flash | Cloudflare Workers AI | October 9 blog: $0.24 and $0.038 per million input tokens. The October 1 cards listed Clef-flash at $0.09. Clef. |
| Clef-omni | Cloudflare Workers AI | $0.15 per million input tokens. BANKING77 macro-F1 94.8. Clef-omni. |
| mammoth-1 | Baseten, POST /v1/systemone | Card Decision Index 0.4 Full 71.24. Live 0.4.1 Full 68.4 rank 3, opened October 23. GET /v1/models with no key returned HTTP 401. mammoth-1. |
| drex-max | Nace, POST /v1/systemone | Decision Index 0.3.2 public 67.59 against Jev 1.14 at 60.70, opened October 20. $0.06 per million input tokens. GET /v1/models with no key returned HTTP 401. Drex Max. |
| d1-ultra | Liquid, POST /decisions/v1/systemone | Artificial Analysis Decision Index 65 against Jev 1.14 at 57, opened October 16. $0.042 per million input tokens. d1-ultra. |
| celeris-1-decision | Celeris, POST /v1/systemone | 80.0% on 3,210 jev-bench items at a 67 ms median, $0.04 per million input tokens. Artificial Analysis Decision Index 52, opened October 16. Celeris-1 Decision. |
| subq-snap-1.1 | Subquadratic, POST /v1/systemone | Artificial Analysis Decision Index 65, tied with d1-ultra and Elephant 9B. $0.04 per million input tokens. Date headers 02:23:17 GMT to 02:23:20 GMT October 23, three POSTs returned HTTP 200. SubQ Snap. |
| pplx-decider-v1.2 | Perplexity, POST /v1/decisions | Decision Index 0.3 overall 68.09 on joint requests. $0.02 per million input tokens. pplx-decider. |
| kettei-fat | Sqwish, POST /v1/decide | JevBench API composite 77.8, rank 1 of 38, opened October 19. $0.06 per million input tokens. kettei-fat. |
| ai_decide | Databricks SQL and REST | Databricks posted October 16 that TypeSafe's Jev-1.14 is available through ai_classify and ai_decide. The docs do not print a price. ai_decide. |
Open checkpoints
Weights that return a choice, a score, or a yes/no. The catalog has the longer table. Repos lists programs that only wrap another server.
| Checkpoint | License and base | One figure from its own page |
|---|---|---|
| basal-omni-1.0-ultra | Apache 2.0. Gemma 4 31B. | JevBench Capability 82.5, first of 184 open systems, opened October 19. Werdykt v1 0.879 on the card. basal-omni. |
| StartLux-Decision v1.1 27B Instant | Code Apache 2.0. Weights CC BY-NC 4.0. | Decision Index 0.3.2 Full score 69.12, first of 174, README opened October 20. StartLux-Decision. |
| Sieve-31B | Apache 2.0. LoRA on Gemma 4 31B. | Decision Index 0.3.2 Full score 63.8, opened October 20. Sieve-31B. |
| Span-02 | Gemma 4 license (Apache 2.0). Gemma 4 26B-A4B. | Decision Index 0.3 text index 63.62. Thinking off 61.2 against Jev 1.13.0 at 58.0. Span-02. |
| Clef, Clef-flash | Apache 2.0. Qwen3.8-27B and Qwen3.5-9B. | Weights are public. Clef. |
| PurpleMIST-Flash-1.0 | Apache 2.0. Qwen3.5-9B-Base, text only. | 0.668 on 15,050 translated African-language decisions. PurpleMIST. |
| Vega 0.8B, vegaml | Apache 2.0. Frozen Qwen3.5-0.8B. | 75.4% on 800 phishing emails, against Jev at 61.9%. vegaml. |
| Laya | Apache 2.0. Encoder, three checkpoints and a router. | JevBench 70.1. Laya. |
| Bespoke Nimble 9B | LoRA on Qwen. The 324-row holdout is the author's labels. | 292 of 324 against Jev 1.13.0 at 302 of 324. Nimble. |
| imajev-4b | Apache 2.0. LoRA on Qwen3.5-4B. Up to two photos. | Opened October 8, 2026: JevBench v1.6.1 capability rank 31 at 61.9. Image JevBench v0.3.0 lists capability 68.0. imajev-4b. |
| Drex DLM | CC BY-NC 4.0. NVIDIA Efficient-DLM-8B. | Decision Index 0.2 cell 52.31. Drex DLM. |
| PolicyLM-1.7B | Apache-2.0. A score from 0 to 1 per category. | Custom-policy accuracy 0.842. PolicyLM. |
Notes
OR-decision-4B prints JevBench 193 of 231, and Choice 79.7 against Jev 79.6 on alphaXiv's chart
alphaXiv posted OR-decision-4B on October 30, 2026, at 16:06 UTC. The Hugging Face card prints JevBench 193 of 231, 83.55%. The launch chart prints Choice 79.7 against Jev 79.6, with sarcasm at 53.6 against Jev 70.7. Apache-2.0 LoRA on Qwen3.5-4B. This desk did not run the weights.
BranchOne returns Choice probabilities from Qwen3.5-0.8B-Base, median 0.305 s on ten short text questions
Necati Demir posted BranchOne on October 29, 2026. The open checkpoint returns one probability per supplied candidate from Qwen3.5-0.8B-Base plus LoRA. His A100 Colab run prints median 0.305 s on ten short text questions. Apache-2.0. This desk did not run the weights.
MiniCPM5-2B-Jev prints 78.79% on JevBench, first among open models at or under 2B
OpenBMB posted MiniCPM5-2B-Jev on October 28, 2026, at 13:01 UTC. The project prints 78.79% on JevBench, first among open models at or under 2B, and 86.41% accuracy with ECE 0.0307 on decision-v7. Apache-2.0 on MiniCPM5-2B. This desk did not run the weights.
Opened October 26, Decision Index 0.4.1 Full score for d3-ultra is 70.18, second of 229
Xunzhuo Liu posted d3-ultra on October 27, 2026, at 03:01 UTC. Opened October 26, Decision Index 0.4.1 Full score is 70.18, second of 229, +3.03 over d3 at 67.15. The card's private SR-Text mean is 87.2. Apache-2.0. This desk did not run the weights.
Decision Model Factory prints DF-JA-Large at 67.63 on S1MB Japanese Task Avg against a public snapshot high of 60.05
Yuma Ichikawa posted Decision Model Factory on October 27, 2026, at 03:19 UTC. arXiv 2610.34956 prints DF-JA-Large at 67.63 on S1MB Japanese Task Avg against a fixed public snapshot high of 60.05. No public weights on the pages this desk opened. This desk did not run a build.
hunch runs EmbeddingGemma2 as a CPU System One endpoint at 20 ms on text and 100 ms on images
Matt Mastracci's hunch turns EmbeddingGemma2 into a CPU System One endpoint at 20 ms on text and 100 ms on images. Cloudflare Clef teaches a local student in real time. Open-source launch was planned for later the same week. This desk has no public repo or hosted call yet.
The full catalog
Hosted calls and open checkpoints are on one page. Each row links the write-up that printed the figure.
Decision models