Identity
- Status
- active
- Developer ID
jev-1.13.0- Released
- Sep 15, 2026
- Version
- Jev
- License
- Unknown
TypeSafeJev
Jev is TypeSafe's first System One model for evaluating typed questions over unstructured state and returning structured decisions with calibrated confidence rather than generated text.
- Token context
- 64K tokens
- Inputs
- Text, Model-specific input
- Outputs
- Model-specific input
- Released
- Sep 15, 2026
Providers & Pricing
Estimate workload cost →| Provider | Input / 1M | Output / 1M | Cache read / 1M | Cache write / 1M | Observed | Evidence |
|---|---|---|---|---|---|---|
TypeSafeAvailable | $0.042 | $0.000 | — | — | Sep 17, 2026 | Provider-reported |
At a Glance
Model Facts
Capacity
- Token context
- 64K tokens
- Maximum output
- Unknown
- Total parameters
- Unknown
- Active parameters
- Unknown
- Knowledge cutoff
- Not reported
Interface and Access
- Inputs
- Text, Model-specific input
- Outputs
- Model-specific input
- API available
- Yes
- Open weights
- No
- Self-hostable
- No
- Reasoning
- No
- Vision
- No
- Tool calling
- No
- Capabilities
- calibrated-confidence, parallel-evaluation, structured outputs, typed-decisions
Technical Specifications
- Combined state and questions limit
- 64,000 tokens
- State plus longest question limit
- 32,000 tokens
- Maximum Choice cardinality
- 255 options
- Default token rate limit
- 250,000 tokens per second
- Default request rate limit
- 1,200 requests per minute
This page represents one developer model product. Dated API snapshots, serving endpoints, pricing, regions, quantizations, and service tiers attach as versioned aliases or provider details and never create duplicate public model records.
Version Lineage
PredecessorUnknown
SuccessorsUnknown
Aliasesjev-1.13.0, jev-latest, jev-preview
Explicit Unknowns
- Maximum output
- Tokenizer
- Knowledge cutoff
Primary Evidence
Sources and Observation Date
Models | TypeSafe AITypeSafe · observed Sep 17, 2026Models and pricing | TypeSafe AITypeSafe · observed Sep 17, 2026Introducing System One Models and Jev | TypeSafeTypeSafe · observed Sep 17, 2026Introduction | TypeSafe AITypeSafe · observed Sep 17, 2026Jev 1.13 Model Jaggedness | TypeSafe AITypeSafe · observed Sep 17, 2026Master Customer Agreement | TypeSafeTypeSafe · observed Sep 17, 2026
Comparable Peers
Related Model Comparisons
| A | Pair | B | Context |
|---|---|---|---|
JevTypeSafe | vs |  GPT-6 AstraOpenAI | cross-developer peerstext |
JevTypeSafe | vs | Claude Fable 5.1Anthropic | cross-developer peerstext |
JevTypeSafe | vs |  Gemini 3.1 ProGoogle DeepMind | cross-developer peerstext |
JevTypeSafe | vs |  DeepSeek V4.1 FlashDeepSeek | cross-developer peerstext |
JevTypeSafe | vs | Grok 4.6xAI | cross-developer peerstext |
JevTypeSafe | vs |  Qwen3.8 MaxQwen | cross-developer peerstext |
JevTypeSafe | vs |  Kimi K3Moonshot AI | cross-developer peerstext |
JevTypeSafe | vs |  MiniMax M3MiniMax | cross-developer peerstext |
JevTypeSafe | vs |  GLM 5.3Z.ai | cross-developer peerstext |
JevTypeSafe | vs |  Hy4 previewTencent | cross-developer peerstext |
JevTypeSafe | vs | Seed 2.1 ProByteDance Seed | cross-developer peerstext |
JevTypeSafe | vs |  Mistral Large 3Mistral AI | cross-developer peerstext |
JevTypeSafe | vs | cross-developer peerstext | |
JevTypeSafe | vs |  Nova 2 LiteAmazon | cross-developer peerstext |
JevTypeSafe | vs | cross-developer peerstext | |
JevTypeSafe | vs |  ERNIE 5.1Baidu | cross-developer peerstext |
JevTypeSafe | vs | cross-developer peerstext | |
JevTypeSafe | vs | cross-developer peerstext | |
JevTypeSafe | vs |  Phi-4 ReasoningMicrosoft | cross-developer peerstext |
JevTypeSafe | vs | cross-developer peerstext | |
JevTypeSafe | vs |  Sonar ProPerplexity | cross-developer peerstext |
JevTypeSafe | vs | Claude Mythos 5.1Anthropic | cross-developer peerstext |
Questions
Jev FAQs
What is Jev?+
Jev is TypeSafe's first System One model for evaluating typed questions over unstructured state and returning structured decisions with calibrated confidence rather than generated text. It is developed by TypeSafe and its current sourced lifecycle status is active.
What inputs and outputs does Jev support?+
Jev's current record lists Text and Model-specific input as input and Model-specific input as output.
How much context does Jev support?+
The current record lists a 64K-token context window and does not report a maximum output length.
What technical specifications are published for Jev?+
Jev's developer-published specifications include combined state and questions limit: 64000 tokens; state plus longest question limit: 32000 tokens; maximum choice cardinality: 255 options; default token rate limit: 250000 tokens per second; default request rate limit: 1200 requests per minute.
How large is Jev?+
The current record lists an unknown total parameter count and an unknown active parameter count. Its architecture is System One parallel-sampler decision model.
What capabilities and tasks does Jev support?+
Jev's recorded capabilities are calibrated-confidence, parallel-evaluation, structured outputs, and typed-decisions. Its supported tasks are classification, extraction, guardrails, routing, and scoring.
Is Jev available through an API?+
Jev is marked as API-available. The directly sourced provider records currently include TypeSafe.
How much does Jev cost through an API?+
The lowest directly sourced prices currently attached to Jev are $0.042 per million input tokens through TypeSafe and $0.000 per million output tokens through TypeSafe. Prices are provider-specific and should be checked against each cited observation date.
Can Jev be self-hosted?+
Jev is not marked as an open-weight model. Self-hosting is marked as unsupported, with no license recorded.
Does Jev have related versions or aliases?+
No predecessor is recorded. No successor is recorded. Recorded aliases are jev-1.13.0, jev-latest, and jev-preview.