DSPy History
https://dspy.ai/api/primitives/History/
dspy.History primitive — typed field holding messages: list[dict] that slots into any Signature . No persistent memory of its own; purely a structured context-injection contract. DSPy's optimisation loop (MIPRO, BootstrapFewShot) treats historical turns as trainable few-shot structure.
Active · Verified 2026-10-22 · fresh as of 2026-10-22 · 45 cited facts, 6 with a third-party source · 2 recorded changes · Suggest a correction
At a glance
- Type
- Signature-field / prompt-injection contract vendor fix
- Tier
- T3
- Section
- Framework-embedded memory
- Created
- 2023-10 (arXiv:2310.03714 posted Oct 2023; grew from DSP framework research started Feb 2022) paper fix
- Latest release
- not applicable — not OSS fix
- GitHub
- stanfordnlp/dspy — ~36.7k stars, last commit n/a repo fix
- Pricing
- OSS free (Apache-2.0); dspy.ai commercial services ~$110K ARR but pricing details not publicly listed third-party fix
- Funding
- Stanford NLP academic project (Hazy Research group); Omar Khattab also affiliated with Databricks third-party fix
- Freshness
- 2026-10-22 fix
- Layer
- agent-infra repo fix
- Licensing
- open-source repo fix
- Developer activity
- OpenRank 14.62 (12-mo avg 16.61) · 103 participants — 2026-09 third-party fix
Memory taxonomy
- storage
- kv
- retrieval
- injection
- persistence
- session
- update
- read-only
- unit
- document
- governance
- inspectable
- conflict
- n/a
When to use
Optimised for: typed Signature-field history primitive
Anti-fit: not for non-DSPy programs
Pros & cons
Pros
Typed programming model means memory composes cleanly with prompts and optimizers.
Cons
DSPy's broader adoption is still emerging; more research-tilted than production-tilted.
Claims & capabilities
Open source. Official Mem0+DSPy tutorial.
Technical surface
- API surface
- not applicable — research paper
- Backend storage
- not applicable — research paper
- Deployment
- Both (OSS self-hosted; dspy.ai cloud services if applicable)
- Embedding model
- not applicable — research paper
- Multi-tenancy
- not applicable — research paper
- MCP
- searched not found
- A2A
- not supported
- OpenTelemetry
- via adapter — Arize Phoenix
Alternatives to DSPy History → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other framework-embedded memory in the catalog, ranked by inbound references.
- Spring AI ChatMemory T3
Wraps conversation memory behind an Advisor abstraction. MessageWindowChatMemory default 20-msg sliding window (preserves system messages); pluggable ChatMemoryRepository backends (in-memory, JDBC, Cassandra, Neo4j). Memory injected as cross-cutting concern at the ChatClient level.
- Flowise Memory T1
Discontinued: Flowise (acquired by Workday, Aug 2025) was sunset in 2026 — code freeze 2026-07-29, GitHub repo archived 2026-08-13, end of life 2026-08-31. Was a visual LangChain canvas with LangChain-native memory nodes (Buffer, Buffer Window, Conversation Summary Buffer) backed by an internal chat_message DB table; Buffer Window kept the last K interactions, Summary Buffer compressed older turns while keeping recent ones verbatim.
- LangGraph Persistence T2
Distinct from LangMem. Built-in checkpointer saves graph state per super-step (short-term, thread-scoped). Store System adds long-term hierarchical key-value memory across threads with optional vector search + TTL. Postgres / Mongo / Redis stores for production.
- AutoGen Memory T2
AutoGen's memory layer: ListMemory chronological context plus teachable agents that vectorise corrections; integrates with Mem0/Zep rather than building deep memory natively. Microsoft AutoGen is in maintenance mode (community managed, bug and security fixes only), and Microsoft points new projects to Microsoft Agent Framework.
- CrewAI Memory T2
Memory subsystem inside the CrewAI orchestration framework; integrates with Mem0 for the long-term tier.
- Hermes Agent memory (Nous Research) T3
Nous Research's open-source self-improving agent keeps an agent-curated MEMORY.md/USER.md notebook injected at session start, full-text (FTS5) search over past sessions that returns raw messages without LLM summarisation, and optional Honcho user modelling. one of the most-starred agent projects of 2026.
Sources (44 cited facts, 6 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Signature-field / prompt-injection contract vendor
- What it does & what's distinct dspy.History primitive — typed field holding messages: list[dict] that slots into any Si… vendor
- Claims / benefits Open source. Official Mem0+DSPy tutorial. vendor
- Modalities text vendor
- Created 2023-10 (arXiv:2310.03714 posted Oct 2023; grew from DSP framework research started Feb … paper
- License MIT repo
- GitHub stanfordnlp/dspy — ~36.7k stars, last commit n/a repo
- Funding Stanford NLP academic project (Hazy Research group); Omar Khattab also affiliated with D… third-party
- Pricing OSS free (Apache-2.0); dspy.ai commercial services ~$110K ARR but pricing details not pu… third-party
- Data handling Library runs in-process; data handling entirely under deployer control; no telemetry repo
- Deployment Both (OSS self-hosted; dspy.ai cloud services if applicable) vendor
- HQ US (Stanford, CA — Stanford NLP / Hazy Research; Omar Khattab also at MIT as of 2025) third-party
- Founders / pedigree Omar Khattab (Stanford PhD, advised by Matei Zaharia & Christopher Potts; now at MIT) third-party
- Memory volume / scale no scale claim vendor
- Versioning none vendor
- Schema evolution free-form / no schema vendor
- A2A support not supported vendor
- OTel via adapter — Arize Phoenix vendor
- Webhooks / events no vendor
- Import / export JSON dspy.save / load vendor
- Integration count 0 vendor
- Orchestration single-agent vendor
- Programmatic control DSL (Signature primitive — typed field) vendor
- Pricing specifics OSS (Apache-2.0) free; dspy.ai commercial services pricing not publicly listed third-party
- Agent abstraction state-machine (Module compose) + ReAct vendor
- Memory primitives thread/conversation (messages: list[dict]) vendor
- LLM lock any (BYO via dspy.LM) vendor
- Runtimes Python vendor
- Session handling stateless (DSPy compiles forward; memory injected as field) vendor
- Validated verticals research, agent developers using DSPy compilers vendor
- Time to running <10min (pip install dspy) vendor
- Anti-fit not for non-DSPy programs vendor
- Optimised for typed Signature-field history primitive vendor
- Adjacent infrastructure requires DSPy; BYO LLM vendor
- Pros Typed programming model means memory composes cleanly with prompts and optimizers. vendor
- Cons DSPy's broader adoption is still emerging; more research-tilted than production-tilted. vendor
- Project & sources dspy History API vendor
- download-trajectory [{"ym":"2025-11","monthly":1217147,"cum":1217147},{"ym":"2025-12","monthly":2495514,"cum… registry
- architecture-deployment-locus local vendor
- licensing-bucket open-source repo
- positioning-platform-vs-product platform vendor
- taxonomy-three-layer-rollup agent-infra repo
- developer-activity-opendigger {"activity":254.56,"as_of":"2026-09","openrank_12mo":16.61,"openrank_current":14.62,"par… third-party
- sdk-languages python vendor
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-23
- Funding / ownership “Stanford NLP academic project (Hazy Research group); Omar Khattab also affiliated with Databricks; dspy.ai commerci…” → “Stanford NLP academic project (Hazy Research group); Omar Khattab also affiliated with Databricks” source
- 2026-10-22
- License “not applicable — not OSS” → “MIT” source
Related systems
References (1)
- Mem0 integrates with — Official Mem0+DSPy tutorial.