---
type: "post"
id: "dspy-3-4-0-out-with-1-native-4439f5"
slug: "dspy-3-4-0-out-with-1-native-4439f5"
title: "DSPy 3.4.0 out, with: (1) native support for Jev and System One models - in the timeless DSPy syntax."
description: "DSPy 3.4.0 moves probability-bearing decisions closer to the program interface, making calibration a first-class concern for agent workflows rather than an external post-processing step."
retrieval_nugget: "DSPy 3.4.0 moves probability-bearing decisions closer to the program interface, making calibration a first-class concern for agent workflows rather than an external post-processing step."
published_at: "2026-09-25"
updated_at: "2026-09-25"
record_date: "2026-09-25"
date_kind: "published_at"
topics: []
source_urls: ["https://github.com/stanfordnlp/dspy/releases/tag/3.4.0"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:dspy-3-4-0-out-with-1-native-4439f5"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:bba6f63ef96d381924457c5488267c982504a6058da201fb652fccc2f5db27a1","reviewed_at":"2026-09-25T20:05:00.485Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":2,"watch_condition_count":2,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"DSPy 3.4.0 moves probability-bearing decisions closer to the program interface, making calibration a first-class concern for agent workflows rather than an external post-processing step.","observed_facts":[{"text":"The release introduces Jev through the TypeSafe client and defines three experimental decision types: Noul for Boolean decisions, Choice for selecting among described alternatives, and Score for ordered rubrics.","source_urls":["https://github.com/stanfordnlp/dspy/releases/tag/3.4.0"]},{"text":"The release notes say ReAnchor fits Boolean thresholds, Score cuts, and Choice weights in Predict programs without rewriting instructions, criteria, or demonstrations.","source_urls":["https://github.com/stanfordnlp/dspy/releases/tag/3.4.0"]},{"text":"DSPy identifies 3.4 as its LM transition release and says 3.5 is the migration deadline for the replaced experimental LM types.","source_urls":["https://github.com/stanfordnlp/dspy/releases/tag/3.4.0"]}],"mechanism":"The mechanism is local decision policy layered over a DSPy predictor. Thresholds, score cuts, and choice weights are applied locally, which the release says allows identical backend requests to reuse cached evidence.","why_now":"This matters now because DSPy calls version 3.4 its LM transition release and names 3.5 as the migration deadline.","implications":["The implication is that teams can separate an underlying model’s evidence from the operational rule that turns that evidence into routing or escalation behavior.","Existing DSPy users with legacy custom-LM integrations or OpenAI-style messages= calls have a defined migration window but should test compatibility before 3.5."],"evidence_boundary":"Evidence boundary: the release notes describe experimental APIs and do not establish that these outputs are statistically calibrated in every setting.","watch_conditions":["Watch whether ReAnchor improves held-out operational metrics across real decision workflows, rather than only fitting thresholds on a development dataset.","Watch whether TypeSafe/Jev support expands beyond top-level decision outputs and the currently unsupported streaming, RLM decision-output, and generative optimization paths."],"related_records":[],"new_branch_reason":"No prior New Runtime record was supplied; this release starts a distinct record because it combines a new decision backend, calibration optimizer, and explicit LM migration milestone."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/dspy-3-4-0-out-with-1-native-4439f5/","markdown":"https://newruntime.com/posts/dspy-3-4-0-out-with-1-native-4439f5.md","json":"https://newruntime.com/posts/dspy-3-4-0-out-with-1-native-4439f5.json"}
---

# DSPy 3.4.0 out, with: (1) native support for Jev and System One models - in the timeless DSPy syntax.

## Retrieval answer

DSPy 3.4.0 moves probability-bearing decisions closer to the program interface, making calibration a first-class concern for agent workflows rather than an external post-processing step.

DSPy 3.4.0 moves probability-bearing decisions closer to the program interface, making calibration a first-class concern for agent workflows rather than an external post-processing step.

The release introduces Jev through the TypeSafe client and defines three experimental decision types: Noul for Boolean decisions, Choice for selecting among described alternatives, and Score for ordered rubrics. The release notes say these types return a decision together with probability or confidence-related evidence. ReAnchor Calibrates Decisions Against Your Metric. Its stated job is to fit Boolean thresholds, Score cuts, and Choice weights in `Predict` programs without rewriting instructions, criteria, or demonstrations.

The mechanism is local decision policy layered over a DSPy predictor. Thresholds, score cuts, and choice weights are applied locally, which the release says allows identical backend requests to reuse cached evidence. This matters now because DSPy calls version 3.4 its LM transition release and names 3.5 as the migration deadline. The implication is that teams can separate an underlying model’s evidence from the operational rule that turns that evidence into routing or escalation behavior. Evidence boundary: the release notes describe experimental APIs and do not establish that these outputs are statistically calibrated in every setting. Watch whether ReAnchor improves held-out operational metrics across real decision workflows, rather than only fitting thresholds on a development dataset.

Source: https://github.com/stanfordnlp/dspy/releases/tag/3.4.0

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)