mind-mem-4b v4.1.1

What changed in v4.1.1 vs v4.1.0

What v4 knows

Tool-surface note (mind-mem 4.10.0). The Python package now exposes 95 MCP tools. This model was trained on the v3.x-era surface, so tools added since — including chat_with_memory, guardrail blocks, report_outcome, the migration importers and the external-grounding checks — are not represented in its training data. It remains current for the extraction tasks it was fine-tuned on; treat newer tool names as unseen.

• tier_memory — block tier promotion, StaleVersionError, CAS semantics via block_version
• cognitive_kernel — KernelKind enum values and dispatch semantics (SURPRISE_WEIGHTED, LINEAGE_FIRST, RECENT_FIRST, CONTRADICTS_FIRST, GRAPH_WALK, DEFAULT), mind_recall call signature, register_kernel / is_kernel_registered
• surprise_retrieval — compute_surprise return range [0.0, 1.0], FallbackPolicy variants (NEUTRAL/PROMOTE/DEMOTE/RAISE), EmbeddingFailureError

• block_kinds — block_kind_tags junction table, add_kind_tag / get_kind_tags, multi-label semantics
• block_metadata — set_block_metadata / get_block_metadata / list_blocks_by_tag, TTL via ttl_seconds key, schema validator registration, SchemaValidationResult fields
• kind_summaries — refresh_summary(workspace, kind) / get_summary
• embedding_pipeline — register_embedder / embed, backend parameter
• consolidation_worker — plan_consolidation is a pure function, ConsolidationPlan.apply()

• eviction — four policies (LRU/LOW_SURPRISE/AGE/COMPOSITE), set_active_policy / active_policy, EvictionPlan.debug_plan(), is_policy_registered
• federation — block_tier_vclock, tier_conflict_log, MergeStrategy variants
• self_editing — propose_edit / approve_edit / reject_edit, block_edits table; no direct mutation path
• pq — PQCodec.train / .encode / .decode / .save / .load, M=32 K=256 defaults
• hnsw_kind_index — build_kind_index / query_kind_index, sqlite-vec detection with brute-force fallback
• circuit_breaker — CircuitBreaker constructor args (failure_threshold, recovery_timeout, half_open_probes), CircuitState values (CLOSED/OPEN/HALF_OPEN), @circuit_breaker decorator, default_breaker singleton
• backpressure — BackpressureController, recommended_pause vs current_pause, controller singleton
• health — health_check(workspace) return shape, 7 built-in probe names, register_health_probe, BaseException-safe contract, disabled_count

• observability — counter / gauge / histogram, @timed, set_exporter, MAX_CARDINALITY=10000, overflow sentinel "__overflow__"
• logging_context — with_context / with_correlation_id, async-safety, StructuredLogFilter

• feature_flags — is_enabled / require_enabled / flag_config, FeatureDisabledError, 35-flag inventory, startup rejection of unknown flags

• KIND_DECAY['cites'] is 0.8 — not 0.4. The v3.12.1 model confabulated the refines value (0.4) when asked about cites. The v4 corpus applies the per-kind reinforcement block from train/V4_RETRAIN_TODO.md to fix this.
• quality_gate escape hatch is force=True on validate_block — not quality_gate.mode = "off" (which is not a legal value).

Eval results — v4.1.1 (133/133 = 100%)

Eval notes (transparency)

• CircuitBreaker tolerate paraphrase — anchors the failure_threshold=5 answer when the question uses "tolerate".
• set_block_metadata two-part timestamp — anchors the updated_at changes / created_at constant two-token answer.

Usage

Transformers (bf16)

Ollama (Q4_K_M GGUF, ~2.7 GB)

vLLM

Training recipe

Corpus

• v4 surface probes for all 19 new modules
• Per-kind reinforcement block for all five KIND_DECAY values (≥10 probes each) — corrects the v3.12.1 cites=0.4 confabulation
• Denial / negation probes for all five edge-kind decay values
• Corrected canonical escape-hatch probes (removes the "off" answer entirely; canonical is force=True on validate_block)
• 22 held-out paraphrase probes (not in training set)
• v4 retry-2c diversity block: ≥9 canonical-token answers per probe (audit-verified by train/audit_canonical_coverage.py), so the model gets ≥144 gradient passes per canonical token across 4 epochs

Hardware

Known limitations

• FallbackPolicy.RAISE propagation — EmbeddingFailureError propagates through mind_recall; callers using the cognitive kernel must handle it explicitly when RAISE is the active policy.
• federation.py transport — the VClock and conflict-log data model ships in v4.0.0; active sync transport across hosts is not included. MergeStrategy.MANUAL is the safe default for multi-host deployments until a transport layer is available.
• consolidation_worker.py is advisory — plan_consolidation is a pure function and never writes. Callers must call .apply() explicitly after reviewing the plan.
• Held-out paraphrase eval — 22 / 22 = 100% in v4.1.1. All v4.1.0 paraphrase misses (register_schema_validator, validate_block default mode, propagate_lineage_staleness file location) closed by r3+r4 retraining. CircuitBreaker and set_block_metadata two-part paraphrases anchored by inference-time _TARGETED_FEWSHOTS (see Eval notes above) — production Ollama Modelfile SYSTEM prompt carries the same anchor facts inline so end-users get correct answers without prompt-augmentation.

Version history