diff --git a/CLAUDE.md b/CLAUDE.md
index b221019..19c5c5b 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -88,7 +88,7 @@ tests/
   with `_dollars` suffix field names. SDK models use short Python names (`yes_bid`)
   with `validation_alias=AliasChoices("yes_bid_dollars", "yes_bid")` to accept both.
   `CreateOrderRequest` serializes with `_dollars` suffix via `serialization_alias`.
-  Verified against OpenAPI spec v3.13.0 on 2026-04-12.
+  Verified against OpenAPI spec v3.18.0 on 2026-05-18.
 - All prices use `Decimal` via the `DollarDecimal` custom Pydantic type.
 - Auth signing payload: `str(timestamp_ms) + METHOD + path_only` (path from urlparse,
   no query params, no trailing slash).
@@ -116,14 +116,14 @@ tests/
 
 ## Testing
 
-- pytest + pytest-asyncio + respx (httpx mock); 1455 tests collected (1407 passing, 48 skipped without live credentials).
+- pytest + pytest-asyncio + respx (httpx mock); ~1920 tests across unit + contract drift suites.
 - Use `respx.mock` for HTTP mocking. Generate test RSA keys via conftest.py fixtures.
 - New function → write a test. Bug fix → write a regression test. New error path → write a test that triggers it.
 
 ## API Reference
 
-- OpenAPI spec: https://docs.kalshi.com/openapi.yaml (v3.13.0, 90+ endpoints)
-- AsyncAPI spec: https://docs.kalshi.com/asyncapi.yaml (11 WebSocket channels)
+- OpenAPI spec: https://docs.kalshi.com/openapi.yaml (v3.18.0, 85 endpoints)
+- AsyncAPI spec: https://docs.kalshi.com/asyncapi.yaml (13 WebSocket channels)
 - Base URL: https://api.elections.kalshi.com/trade-api/v2
 - Demo URL: https://demo-api.kalshi.co/trade-api/v2
 - Auth: RSA-PSS / SHA256 / MGF1(SHA256) / salt_length=DIGEST_LENGTH / base64
diff --git a/README.md b/README.md
index b75bcca..237954b 100644
--- a/README.md
+++ b/README.md
@@ -11,7 +11,9 @@ A professional, spec-first Python SDK for the [Kalshi](https://kalshi.com) predi
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 [![Type checked: mypy strict](https://img.shields.io/badge/mypy-strict-blue.svg)](https://mypy.readthedocs.io/)
 
-- **Full coverage** of the Kalshi REST API (89 endpoints across 19 resources) and WebSocket API (11 channels).
+- **Full coverage** of the Kalshi REST API (85 endpoints across 19 resources, OpenAPI v3.18.0) and WebSocket API (13 channels).
+- **V2 event-market orders**: `create_v2` / `amend_v2` / `decrease_v2` / `cancel_v2` plus batched variants on `/portfolio/events/orders/*`. Legacy `/portfolio/orders` keeps working — deprecated no earlier than May 6, 2026.
+- **Funding & cost introspection**: `portfolio.deposits()`, `portfolio.withdrawals()`, `account.endpoint_costs()`.
 - **Sync and async** clients sharing one transport — no thread-pool wrapping.
 - **Typed end-to-end**: Pydantic v2 models, `mypy --strict` clean, ships `py.typed`. `Literal` types on fixed-enum kwargs.
 - **Spec-aligned with drift guards**: hard-fail contract tests catch query, body, and WebSocket payload drift on every commit.
@@ -141,6 +143,34 @@ client.orders.create(request=CreateOrderRequest(
 ))
 ```
 
+### V2 event-market orders
+
+Spec v3.18.0 introduced the V2 family on `/portfolio/events/orders/*` —
+event-scoped semantics with single-book `bid`/`ask` sides and fixed-point
+dollar prices. Legacy `/portfolio/orders` keeps working and will be
+deprecated no earlier than May 6, 2026.
+
+```python
+import uuid
+from decimal import Decimal
+from kalshi import KalshiClient, CreateOrderV2Request
+
+with KalshiClient.from_env() as client:
+    resp = client.orders.create_v2(request=CreateOrderV2Request(
+        ticker="EVENT-MKT",
+        client_order_id=str(uuid.uuid4()),  # required + server idempotency key
+        side="bid",                         # BookSideLiteral: "bid" | "ask"
+        count=Decimal("10"),
+        price=Decimal("0.50"),
+        time_in_force="good_till_canceled",
+        self_trade_prevention_type="taker_at_cross",
+    ))
+    print(resp.order_id, resp.remaining_count, resp.fill_count)
+```
+
+The V2 surface is model-only (no kwarg overload); pass a fully-constructed
+request model. See [V2 orders docs](https://texascoding.github.io/kalshi-python-sdk/resources/orders/#v2-event-market-orders) for amend/decrease/batch variants.
+
 ## WebSocket streaming
 
 ```python
@@ -161,9 +191,10 @@ async def main() -> None:
 asyncio.run(main())
 ```
 
-Available channels (11): `ticker`, `trade`, `orderbook_delta`, `fill`,
-`market_positions`, `user_orders`, `order_group`, `market_lifecycle`,
-`multivariate`, `multivariate_lifecycle`, `communications`.
+Available channels (13): `ticker`, `trade`, `orderbook_delta`, `fill`,
+`market_positions`, `user_orders`, `order_group_updates`,
+`market_lifecycle_v2`, `multivariate`, `multivariate_market_lifecycle`,
+`communications`, `control_frames`, `root`.
 
 ## Error handling
 
diff --git a/ROADMAP.md b/ROADMAP.md
index 3a63542..9b6916d 100644
--- a/ROADMAP.md
+++ b/ROADMAP.md
@@ -4,6 +4,15 @@
 
 See `CHANGELOG.md` for full release history.
 
+- **v2.1.0 (2026-05-18)** — OpenAPI sync v3.13.0 → v3.18.0 (`#155`). V2
+  event-market orders family (`create_v2` / `amend_v2` / `decrease_v2` /
+  `cancel_v2` + batched variants on `/portfolio/events/orders/*`),
+  deposit/withdrawal history, account endpoint-cost introspection,
+  `Balance.balance_dollars` (soft-breaking at construction sites only),
+  optional `exchange_index` / `user_filter` / `rfq_user_filter` /
+  `incentive_description` / `post_only` kwargs on existing endpoints.
+  Also fixes a recurring false-alarm in the weekly spec-drift workflow
+  by committing `specs/asyncapi.yaml` as a pinned snapshot.
 - **v2.0.0 (2026-05-17)** — audit-driven hardening. 30 audit findings closed
   across five parallel waves (`#77`–`#106`) plus follow-ups: WebSocket
   recv-loop overhaul (5 reconnect races + narrowed exceptions),
@@ -36,7 +45,13 @@ See `CHANGELOG.md` for full release history.
 
 ## Next milestone
 
-Not scoped. Candidates from the v2.0 audit backlog:
+Not scoped. Carry-overs from v2.1 and v2.0 audit backlog:
+
+- **Response-side spec drift detection.** v2.1 contract tests cover request
+  bodies (`TestRequestBodyDrift`) but not response models, which is how
+  `Balance.balance_dollars` slipped through 5 rounds of review. Add a
+  `RESPONSE_MODEL_MAP` + walker that asserts spec-required fields exist on
+  the SDK model.
 
 - Apply `extra="allow"` policy to WS envelope models (`kalshi/ws/models/`)
   to mirror the response-model uniformity from `#114`.
diff --git a/docs/concepts.md b/docs/concepts.md
index a2abd9c..0c1c19e 100644
--- a/docs/concepts.md
+++ b/docs/concepts.md
@@ -66,6 +66,14 @@ anything with `_dollars` or `yes_price` / `no_price` is `Decimal` dollars.
 - **Position** — your aggregate exposure on a market (signed by side).
 - **Settlement** — what the exchange paid out when the market resolved.
 
+Orders come in two families:
+
+- **V1** — `/portfolio/orders/*`. Yes/no sides, paired `yes_price` /
+  `no_price`. Stable surface; deprecation no earlier than May 6, 2026.
+- **V2** — `/portfolio/events/orders/*`. Event-scoped, single-book
+  `bid` / `ask` sides, single `price` field, required `client_order_id`
+  acting as an idempotency key. Use this for new event-market integrations.
+
 See [Orders](resources/orders.md), [Portfolio](resources/portfolio.md).
 
 ## RFQ and Quote
diff --git a/docs/index.md b/docs/index.md
index 985fa19..127bbb6 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -3,9 +3,14 @@
 A professional, spec-first Python SDK for the [Kalshi](https://kalshi.com) prediction
 markets API.
 
-- **Full REST coverage** — 89 endpoints across 19 resources, every kwarg drift-tested
-  against the OpenAPI spec.
-- **Full WebSocket coverage** — 11 channels with sequence-gap detection, automatic
+- **Full REST coverage** — 85 endpoints across 19 resources (OpenAPI v3.18.0),
+  every kwarg drift-tested against the spec.
+- **V2 event-market orders** — new `create_v2` / `amend_v2` / `decrease_v2` /
+  `cancel_v2` family on `/portfolio/events/orders/*`. Legacy `/portfolio/orders`
+  keeps working; deprecation no earlier than May 6, 2026.
+- **Funding + cost introspection** — `portfolio.deposits()`,
+  `portfolio.withdrawals()`, `account.endpoint_costs()`.
+- **Full WebSocket coverage** — 13 channels with sequence-gap detection, automatic
   reconnection, backpressure strategies, and an in-memory orderbook builder.
 - **Sync and async parity** — `KalshiClient` and `AsyncKalshiClient` share one
   transport implementation; method names, kwargs, return types, and error behavior
diff --git a/docs/migration.md b/docs/migration.md
index 381cf77..9bb7c66 100644
--- a/docs/migration.md
+++ b/docs/migration.md
@@ -1,5 +1,112 @@
 # Migration
 
+## v2.0 → v2.1
+
+v2.1 syncs the SDK to OpenAPI spec v3.18.0. It's **additive at the resource
+surface** — eight new endpoints, several new optional kwargs on existing
+methods, and one soft-breaking model-construction change called out below.
+Code that only consumes the SDK's responses needs no edits; code that
+constructs models directly in tests/mocks needs one small update.
+
+### `Balance.balance_dollars` — required, soft-breaking at construction
+
+Spec v3.18.0 adds `balance_dollars: FixedPointDollars` to
+`GetBalanceResponse` as a required field. The server now guarantees it, so
+callers parsing API responses (`client.portfolio.balance()`) are unaffected.
+**But** any code that builds `Balance(...)` directly — typically test mocks
+— will hit `ValidationError` until it adds the field.
+
+```python
+# v2.0 — broken in v2.1
+Balance(balance=50000, portfolio_value=75000, updated_ts=ts)
+
+# v2.1
+Balance(
+    balance=50000,
+    balance_dollars=Decimal("500.00"),   # new required field
+    portfolio_value=75000,
+    updated_ts=ts,
+)
+```
+
+The accompanying optional `balance_breakdown: list[IndexedBalance] | None`
+splits the total across exchange shards when present. `IndexedBalance.balance`
+is `DollarDecimal` (matching `balance_dollars` units), not cents — same
+field name as `Balance.balance` but a different type. Be deliberate when
+reading from `balance.balance_breakdown[i].balance`.
+
+### V2 event-market orders
+
+Six new methods on `OrdersResource` / `AsyncOrdersResource` hit the new
+`/portfolio/events/orders/*` paths:
+
+- `create_v2(*, request: CreateOrderV2Request)`
+- `cancel_v2(order_id, *, subaccount, exchange_index)`
+- `amend_v2(order_id, *, request: AmendOrderV2Request, subaccount)`
+- `decrease_v2(order_id, *, request: DecreaseOrderV2Request, subaccount)`
+- `batch_create_v2(*, request: BatchCreateOrdersV2Request)`
+- `batch_cancel_v2(*, request: BatchCancelOrdersV2Request)`
+
+Legacy `/portfolio/orders` keeps working and will be deprecated no earlier
+than May 6, 2026. No migration is required to stay on v1 paths. New
+event-market trading should target the V2 family for the cleaner shape
+(single `bid`/`ask` side, single `price` field, explicit idempotency).
+
+Two important differences from V1:
+
+1. **`client_order_id` is required** on `CreateOrderV2Request` and acts as
+   the server-side idempotency key. Reusing a value returns the original
+   order rather than placing a new one. Use a fresh UUID4 per call.
+2. **`side` uses `BookSideLiteral` (`"bid"` / `"ask"`)**, not V1's
+   `SideLiteral` (`"yes"` / `"no"`).
+
+### Spec-driven asymmetry on V2 amend/decrease
+
+`amend_v2` and `decrease_v2` accept `subaccount` as a **resource-method
+kwarg** (query param on the wire) but read `exchange_index` from the
+**request body**. `cancel_v2` differs again — both are query params there,
+because that endpoint has no body. This mirrors the spec exactly:
+
+```python
+client.orders.amend_v2(
+    "ord-1",
+    subaccount=3,                          # query param
+    request=AmendOrderV2Request(
+        ticker="EVENT-MKT", side="bid",
+        price=Decimal("0.55"),
+        count=Decimal("10"),
+        exchange_index=0,                  # body field
+    ),
+)
+```
+
+### New optional kwargs on existing endpoints
+
+All additive — existing call sites keep working:
+
+- `orders.cancel(*, exchange_index)`, `order_groups.delete(*, exchange_index)`
+- `communications.list_rfqs(*, user_filter)`, `communications.list_all_rfqs(*, user_filter)`
+- `communications.list_quotes(*, user_filter, rfq_user_filter)`, `communications.list_all_quotes(*, user_filter, rfq_user_filter)`. `user_filter="self"` and `rfq_user_filter="self"` are now standalone satisfiers for the server-side filter requirement (previously only `quote_creator_user_id` / `rfq_creator_user_id` worked).
+- `incentive_programs.list(*, incentive_description)` and the `list_all` variant.
+- `CreateQuoteRequest.post_only` — pass `post_only=True` to `create_quote()`.
+- `exchange_index` on order/amend/decrease/batch-cancel request models.
+
+### New endpoints (additive)
+
+- `portfolio.deposits()` / `portfolio.deposits_all()` — deposit history.
+- `portfolio.withdrawals()` / `portfolio.withdrawals_all()` — withdrawal history.
+- `account.endpoint_costs()` — token costs for endpoints whose cost differs
+  from the default.
+
+### New public types
+
+Added to `kalshi.*` and `kalshi.models.*`:
+
+- Literals: `BookSideLiteral`, `UserFilterLiteral`, `PaymentStatusLiteral`, `PaymentTypeLiteral`.
+- Models: `Deposit`, `Withdrawal`, `IndexedBalance`, `AccountEndpointCosts`, `EndpointTokenCost`, and the V2 request/response family.
+
+---
+
 ## v1.x → v2.0
 
 v2.0 is mostly additive (new `max_pages` kwarg, `KalshiConfig.http2`/`limits`,
diff --git a/docs/request-models.md b/docs/request-models.md
index bedda48..e03109f 100644
--- a/docs/request-models.md
+++ b/docs/request-models.md
@@ -64,6 +64,11 @@ OpenAPI spec.
 | `client.orders.batch_cancel` | `BatchCancelOrdersRequest` (wraps `list[BatchCancelOrdersRequestOrder]`) |
 | `client.orders.amend` | `AmendOrderRequest` |
 | `client.orders.decrease` | `DecreaseOrderRequest` |
+| `client.orders.create_v2` | `CreateOrderV2Request` (V2 event-market — model-only, no kwarg form) |
+| `client.orders.amend_v2` | `AmendOrderV2Request` (V2 — model-only) |
+| `client.orders.decrease_v2` | `DecreaseOrderV2Request` (V2 — model-only, XOR `reduce_by`/`reduce_to`) |
+| `client.orders.batch_create_v2` | `BatchCreateOrdersV2Request` (wraps `list[CreateOrderV2Request]`) |
+| `client.orders.batch_cancel_v2` | `BatchCancelOrdersV2Request` (wraps `list[BatchCancelOrdersV2RequestOrder]`) |
 | `client.api_keys.create` | `CreateApiKeyRequest` |
 | `client.api_keys.generate` | `GenerateApiKeyRequest` |
 | `client.communications.create_rfq` | `CreateRFQRequest` |
@@ -94,12 +99,42 @@ an awkward wire name:
 
 You never type these long names — they're an internal implementation detail.
 
+## V2 surface: model-only
+
+The V2 event-market order endpoints (`create_v2` / `amend_v2` / `decrease_v2`
+/ `batch_*_v2`) don't accept individual kwargs — pass a fully-constructed
+request model. This is intentional: V2 took the opportunity to drop the
+V1 kwarg-overload surface entirely, so the model is the single source of
+truth for the payload shape.
+
+```python
+import uuid
+from decimal import Decimal
+from kalshi import CreateOrderV2Request
+
+req = CreateOrderV2Request(
+    ticker="EVENT-MKT",
+    client_order_id=str(uuid.uuid4()),   # required — idempotency key
+    side="bid",
+    count=Decimal("10"),
+    price=Decimal("0.50"),
+    time_in_force="good_till_canceled",
+    self_trade_prevention_type="taker_at_cross",
+)
+client.orders.create_v2(request=req)
+```
+
+`CreateOrderV2Request.client_order_id` is required (V1's was optional) and
+the server treats it as an idempotency key — reusing one returns the
+original order rather than placing a new one. Generate a fresh UUID4 per
+call.
+
 ## Cross-field invariants
 
 Some request models enforce relationships **before** the HTTP call:
 
-- `DecreaseOrderRequest` — exactly one of `reduce_by` / `reduce_to` is
-  required (XOR enforced via a model validator).
+- `DecreaseOrderRequest` / `DecreaseOrderV2Request` — exactly one of
+  `reduce_by` / `reduce_to` is required (XOR enforced via a model validator).
 - `AmendOrderRequest` — at least one of `yes_price` / `no_price` / `count`
   must be present (enforced in the resource method, before the body is built).
 
diff --git a/docs/resources/account.md b/docs/resources/account.md
index d778603..e6f2ea2 100644
--- a/docs/resources/account.md
+++ b/docs/resources/account.md
@@ -9,6 +9,7 @@ Auth required.
 | Method | Endpoint |
 |---|---|
 | `limits()` | `GET /account/limits` |
+| `endpoint_costs()` | `GET /account/endpoint_costs` |
 
 ## Read tier limits
 
@@ -29,6 +30,23 @@ calls.
     the production server returns the nested `RateLimit` objects shown above.
     The SDK normalizes to the live shape.
 
+## Endpoint costs
+
+New in v2.1.0. Lists API v2 endpoints whose configured token cost differs
+from the default — useful for budgeting against your write tier.
+
+```python
+costs = client.account.endpoint_costs()
+print(costs.default_cost)               # int, the baseline (typically 10)
+for entry in costs.endpoint_costs:
+    # EndpointTokenCost: method (str), path (str), cost (int)
+    print(entry.method, entry.path, entry.cost)
+```
+
+Endpoints not present in `endpoint_costs` use `default_cost`. Batch
+endpoints typically appear here with a per-item multiplier (e.g.
+`POST /portfolio/orders/batched` costs ~10 tokens per order in the batch).
+
 ## Reference
 
 ::: kalshi.resources.account.AccountResource
diff --git a/docs/resources/communications.md b/docs/resources/communications.md
index a695529..791b8d8 100644
--- a/docs/resources/communications.md
+++ b/docs/resources/communications.md
@@ -71,10 +71,51 @@ client.communications.confirm_quote(resp.quote.quote_id)
 ```
 
 !!! warning "`list_quotes` requires a user-id filter"
-    `list_quotes` and `list_all_quotes` **must** be called with one of
-    `quote_creator_user_id=` or `rfq_creator_user_id=`. Passing `rfq_id=`
-    alone raises `ValueError` locally before the round trip — this enforces
-    a server-side requirement.
+    `list_quotes` and `list_all_quotes` **must** be called with at least one of:
+
+    - `quote_creator_user_id=` (filter to a specific quoter)
+    - `rfq_creator_user_id=` (filter to a specific RFQ originator)
+    - `user_filter="self"` (server-side shorthand for "the caller's quotes")
+    - `rfq_user_filter="self"` (server-side shorthand for "quotes on the caller's RFQs")
+
+    `user_filter` / `rfq_user_filter` were added in spec v3.18.0 and let
+    you avoid round-tripping `get_id()` first. Passing `rfq_id=` alone
+    raises `ValueError` locally before the round trip — this enforces a
+    server-side requirement.
+
+## Filtering shortcuts (v2.1.0)
+
+```python
+# All quotes you made — no get_id() needed.
+for q in client.communications.list_all_quotes(user_filter="self"):
+    print(q.quote_id, q.yes_bid)
+
+# All quotes against RFQs you originated.
+for q in client.communications.list_all_quotes(rfq_user_filter="self"):
+    ...
+
+# Same shortcut on RFQs:
+for rfq in client.communications.list_all_rfqs(user_filter="self"):
+    ...
+```
+
+`UserFilterLiteral` only accepts `"self"` today — the spec leaves room for
+server-side shorthands like `"organization"` in the future without an SDK
+upgrade.
+
+## Post-only quotes
+
+`create_quote()` accepts `post_only=True` (added in v2.1.0) to ensure your
+resting order is canceled rather than crossed if it would take liquidity:
+
+```python
+client.communications.create_quote(
+    rfq_id="rfq_abc",
+    yes_bid="0.60", no_bid="0.40",
+    rest_remainder=True,
+    post_only=True,    # cancel rather than cross
+)
+```
 
 ## RFQ statuses
 
diff --git a/docs/resources/incentive-programs.md b/docs/resources/incentive-programs.md
index f11e7a9..a013964 100644
--- a/docs/resources/incentive-programs.md
+++ b/docs/resources/incentive-programs.md
@@ -18,13 +18,16 @@ Public — no auth required.
 page = client.incentive_programs.list(
     status="active",                     # IncentiveProgramStatusLiteral
     incentive_type="liquidity",          # IncentiveProgramTypeLiteral — spec "type"
-    series_ticker="KXPRES",
+    incentive_description="Q2 maker",    # new in v2.1.0 — exact-match filter
     limit=100,
 )
 for p in page:
     print(p.program_id, p.title, p.start_ts, p.end_ts)
 ```
 
+`incentive_description` (added in v2.1.0) is an exact-string filter on the
+program's description — useful when multiple programs share a status/type.
+
 Use `list_all` to walk every program:
 
 ```python
diff --git a/docs/resources/order-groups.md b/docs/resources/order-groups.md
index 9ed37b3..6a4c1b3 100644
--- a/docs/resources/order-groups.md
+++ b/docs/resources/order-groups.md
@@ -11,8 +11,8 @@ Auth required throughout.
 |---|---|
 | `list(*, subaccount=None)` | `GET /portfolio/order_groups` |
 | `get(order_group_id, *, subaccount=None)` | `GET /portfolio/order_groups/{id}` |
-| `create(*, contracts_limit, subaccount=None)` | `POST /portfolio/order_groups/create` |
-| `delete(order_group_id, *, subaccount=None)` | `DELETE /portfolio/order_groups/{id}` |
+| `create(*, contracts_limit, subaccount=None, exchange_index=None)` | `POST /portfolio/order_groups/create` |
+| `delete(order_group_id, *, subaccount=None, exchange_index=None)` | `DELETE /portfolio/order_groups/{id}` |
 | `reset(order_group_id, *, subaccount=None)` | `PUT /portfolio/order_groups/{id}/reset` |
 | `trigger(order_group_id, *, subaccount=None)` | `PUT /portfolio/order_groups/{id}/trigger` |
 | `update_limit(order_group_id, *, contracts_limit)` | `PUT /portfolio/order_groups/{id}/limit` |
@@ -67,6 +67,19 @@ client.order_groups.update_limit("og_abc", contracts_limit=200)
 subaccount query parameter on this path. Route via the group's own
 subaccount-on-create instead.
 
+## Exchange index (v2.1.0)
+
+`create()` and `delete()` gained an optional `exchange_index: int | None`
+parameter. Currently only `exchange_index=0` is supported per spec; the
+field is reserved for future multi-shard fanout. Pass it for forward
+compatibility if you're writing infrastructure that may target a non-zero
+shard later:
+
+```python
+client.order_groups.create(contracts_limit=100, exchange_index=0)
+client.order_groups.delete("og_abc", exchange_index=0)
+```
+
 ## Reference
 
 ::: kalshi.resources.order_groups.OrderGroupsResource
diff --git a/docs/resources/orders.md b/docs/resources/orders.md
index 6116e49..87456c6 100644
--- a/docs/resources/orders.md
+++ b/docs/resources/orders.md
@@ -7,13 +7,15 @@ Every method here requires auth.
 
 ## Quick reference
 
+### V1 (legacy `/portfolio/orders/*`) — deprecated no earlier than May 6, 2026
+
 | Method | Endpoint | Retry |
 |---|---|---|
 | `create(...)` | `POST /portfolio/orders` | **never** — see [Retries & idempotency](../retries.md) |
 | `batch_create(orders)` | `POST /portfolio/orders/batched` | never |
 | `get(order_id)` | `GET /portfolio/orders/{order_id}` | yes (GET) |
 | `list(...)` / `list_all(...)` | `GET /portfolio/orders` | yes |
-| `cancel(order_id, *, subaccount=None)` | `DELETE /portfolio/orders/{order_id}` | never |
+| `cancel(order_id, *, subaccount=None, exchange_index=None)` | `DELETE /portfolio/orders/{order_id}` | never |
 | `batch_cancel(orders)` | `DELETE /portfolio/orders/batched` | never |
 | `amend(order_id, ...)` | `POST /portfolio/orders/{order_id}/amend` | never |
 | `decrease(order_id, *, reduce_by, reduce_to)` | `POST /portfolio/orders/{order_id}/decrease` | never |
@@ -21,6 +23,22 @@ Every method here requires auth.
 | `queue_positions(*, market_tickers, event_ticker)` | `GET /portfolio/orders/queue_positions` | yes |
 | `queue_position(order_id)` | `GET /portfolio/orders/{order_id}/queue_position` | yes |
 
+### V2 event-market orders (`/portfolio/events/orders/*`) — recommended for new code
+
+| Method | Endpoint | Retry |
+|---|---|---|
+| `create_v2(*, request)` | `POST /portfolio/events/orders` | never |
+| `batch_create_v2(*, request)` | `POST /portfolio/events/orders/batched` | never |
+| `cancel_v2(order_id, *, subaccount, exchange_index)` | `DELETE /portfolio/events/orders/{order_id}` | never |
+| `batch_cancel_v2(*, request)` | `DELETE /portfolio/events/orders/batched` | never |
+| `amend_v2(order_id, *, request, subaccount)` | `POST /portfolio/events/orders/{order_id}/amend` | never |
+| `decrease_v2(order_id, *, request, subaccount)` | `POST /portfolio/events/orders/{order_id}/decrease` | never |
+
+V2 uses `BookSideLiteral` (`"bid"` / `"ask"`), a single `price` field
+(not paired yes/no), and requires `client_order_id` as a server-side
+idempotency key. The surface is **model-only** — every method takes a
+pre-built request model rather than kwargs.
+
 ## Place an order
 
 ```python
@@ -181,6 +199,140 @@ q = client.orders.queue_position("ord_abc")  # returns Decimal
 `queue_position(order_id)` returns a bare `Decimal`, not a model — that's the
 shape Kalshi's endpoint emits.
 
+## V2 event-market orders
+
+Spec v3.18.0 introduced the `/portfolio/events/orders/*` family. Legacy
+`/portfolio/orders` keeps working and will be deprecated **no earlier than
+May 6, 2026** — start new event-market integrations on V2.
+
+### How V2 differs from V1
+
+| | V1 (`/portfolio/orders`) | V2 (`/portfolio/events/orders`) |
+|---|---|---|
+| Side enum | `SideLiteral` — `"yes"` / `"no"` | `BookSideLiteral` — `"bid"` / `"ask"` |
+| Price | Paired `yes_price` / `no_price` | Single `price: DollarDecimal` |
+| `client_order_id` | Optional | **Required** + acts as server-side idempotency key |
+| Kwarg API | Yes (kwargs or request model) | Model-only — pass a request model |
+| Side filter | n/a | per-spec query/body asymmetry — see [Asymmetry](#asymmetry) |
+
+### Place a V2 order
+
+```python
+import uuid
+from decimal import Decimal
+from kalshi import KalshiClient, CreateOrderV2Request
+
+with KalshiClient.from_env() as client:
+    resp = client.orders.create_v2(request=CreateOrderV2Request(
+        ticker="EVENT-MKT-A",
+        client_order_id=str(uuid.uuid4()),   # required — idempotency key
+        side="bid",                          # BookSideLiteral
+        count=Decimal("10"),
+        price=Decimal("0.50"),
+        time_in_force="good_till_canceled",
+        self_trade_prevention_type="taker_at_cross",
+        # post_only=False, reduce_only=False, cancel_order_on_pause=False,
+        # subaccount=0, order_group_id="og_abc", exchange_index=0,
+    ))
+    print(resp.order_id, resp.remaining_count, resp.fill_count)
+```
+
+Reusing a `client_order_id` for a second call returns the **original**
+order rather than placing a new one. Generate a fresh UUID4 per request.
+
+### Cancel V2
+
+```python
+result = client.orders.cancel_v2("ord_v2_1", subaccount=0, exchange_index=0)
+print(result.reduced_by, result.ts_ms)   # FixedPointCount, int
+```
+
+`cancel_v2` returns a `CancelOrderV2Response` (with the count actually
+canceled), not `None` like the V1 `cancel`. A 204 No Content from the
+server is treated as a protocol violation and raises `KalshiError`.
+
+### Amend V2 / Decrease V2
+
+```python
+from decimal import Decimal
+from kalshi import AmendOrderV2Request, DecreaseOrderV2Request
+
+resp = client.orders.amend_v2(
+    "ord_v2_1",
+    subaccount=0,                              # query param
+    request=AmendOrderV2Request(
+        ticker="EVENT-MKT-A",
+        side="bid",
+        price=Decimal("0.55"),
+        count=Decimal("12"),                   # total/max fillable count
+        exchange_index=0,                      # body field
+    ),
+)
+
+resp = client.orders.decrease_v2(
+    "ord_v2_1",
+    subaccount=0,
+    request=DecreaseOrderV2Request(
+        reduce_by=Decimal("3"),                # exactly one of reduce_by | reduce_to
+    ),
+)
+```
+
+`DecreaseOrderV2Request` enforces XOR on `reduce_by` / `reduce_to` at
+construction — passing both or neither raises `ValidationError`.
+
+### Batch V2
+
+```python
+from kalshi import (
+    BatchCreateOrdersV2Request, BatchCancelOrdersV2Request,
+    BatchCancelOrdersV2RequestOrder,
+)
+
+# Create
+result = client.orders.batch_create_v2(request=BatchCreateOrdersV2Request(
+    orders=[
+        CreateOrderV2Request(...),    # per-order builds, same shape as create_v2
+        CreateOrderV2Request(...),
+    ],
+))
+for entry in result.orders:
+    if entry.error is not None:
+        print("failed:", entry.error)
+    else:
+        print(entry.order_id, entry.fill_count)
+
+# Cancel
+result = client.orders.batch_cancel_v2(request=BatchCancelOrdersV2Request(
+    orders=[
+        BatchCancelOrdersV2RequestOrder(order_id="ord_a", subaccount=0),
+        BatchCancelOrdersV2RequestOrder(order_id="ord_b"),
+    ],
+))
+```
+
+Per-entry error handling is built into `BatchCreateOrdersV2ResponseEntry`
+and `BatchCancelOrdersV2ResponseEntry` — successful entries carry the
+order/fill data, failed entries carry an `error` dict with the rest of the
+fields nulled (create) or zeroed (cancel).
+
+### Asymmetry { #asymmetry }
+
+This trips people up. The V2 spec routes `subaccount` and `exchange_index`
+to **different places** depending on the endpoint:
+
+| Endpoint | `subaccount` | `exchange_index` |
+|---|---|---|
+| `cancel_v2` | query | query |
+| `amend_v2` | query | **body** (on `AmendOrderV2Request`) |
+| `decrease_v2` | query | **body** (on `DecreaseOrderV2Request`) |
+| `create_v2` / `batch_*_v2` | n/a — body on request model | body |
+
+This is faithful to OpenAPI v3.18.0 — `amend_v2`/`decrease_v2` only
+declare `SubaccountQueryDefaultPrimary` in their `parameters` list,
+while `cancel_v2` declares both that and `ExchangeIndexQuery`. Don't try
+to "normalize" it; the SDK matches the spec.
+
 ## Reference
 
 ::: kalshi.resources.orders.OrdersResource
diff --git a/docs/resources/portfolio.md b/docs/resources/portfolio.md
index 12df609..f4102e7 100644
--- a/docs/resources/portfolio.md
+++ b/docs/resources/portfolio.md
@@ -11,15 +11,43 @@ Auth required throughout.
 | `positions(*, ...)` | `GET /portfolio/positions` |
 | `settlements(...)` / `settlements_all(...)` | `GET /portfolio/settlements` |
 | `total_resting_order_value()` | `GET /portfolio/summary/total_resting_order_value` (FCM only) |
+| `deposits(*, limit, cursor)` / `deposits_all(*, limit, max_pages)` | `GET /portfolio/deposits` |
+| `withdrawals(*, limit, cursor)` / `withdrawals_all(*, limit, max_pages)` | `GET /portfolio/withdrawals` |
 
 ## Balance
 
 ```python
 bal = client.portfolio.balance()
-print(bal.balance, bal.portfolio_value, bal.updated_ts)
+print(bal.balance, bal.balance_dollars, bal.portfolio_value, bal.updated_ts)
 ```
 
-`Balance.balance` and `portfolio_value` are **integer cents**, not dollars.
+`Balance.balance` and `portfolio_value` are **integer cents**.
+`Balance.balance_dollars` (required, new in v2.1.0) is the same amount as a
+`DollarDecimal` — use whichever shape your code expects without manually
+dividing by 100.
+
+`Balance.balance_breakdown` (optional, also new in v2.1.0) splits the total
+across exchange shards:
+
+```python
+if bal.balance_breakdown is not None:
+    for shard in bal.balance_breakdown:
+        # IndexedBalance.balance is DollarDecimal, not cents — same name
+        # as Balance.balance but a different type. See note below.
+        print(shard.exchange_index, shard.balance)
+```
+
+!!! warning "Type collision: `Balance.balance` vs. `IndexedBalance.balance`"
+    `Balance.balance` is **integer cents**. `IndexedBalance.balance` (inside
+    `balance_breakdown`) is `DollarDecimal` (dollars), matching
+    `Balance.balance_dollars` units. Same field name, different types — be
+    deliberate when iterating the breakdown.
+
+!!! note "Constructing `Balance` directly"
+    v2.1.0 made `balance_dollars` required to match spec v3.18.0. Existing
+    code that calls `client.portfolio.balance()` is unaffected. Code that
+    builds `Balance(...)` directly (typically in tests/mocks) needs the
+    field added — see [Migration](../migration.md#v20-v21).
 
 ## Positions
 
@@ -85,6 +113,34 @@ print(total.total_value)
     Non-FCM accounts get a `403` (mapped to `KalshiAuthError`). Demo mirrors
     production behavior here.
 
+## Deposits and withdrawals
+
+New in v2.1.0. Standard `Page[T]` pagination — see [Pagination](../pagination.md).
+
+```python
+# Most recent deposits
+page = client.portfolio.deposits(limit=50)
+for d in page:
+    print(d.id, d.status, d.type, d.amount_cents, d.created_ts)
+
+# All deposits
+for d in client.portfolio.deposits_all():
+    ...
+
+# Withdrawals follow the same shape
+for w in client.portfolio.withdrawals_all():
+    print(w.id, w.status, w.amount_cents, w.fee_cents)
+```
+
+`Deposit` and `Withdrawal` are structurally identical: `id`, `status`
+(`PaymentStatusLiteral`: `"pending"` / `"applied"` / `"failed"` /
+`"returned"`), `type` (`PaymentTypeLiteral`: `"ach"` / `"wire"` /
+`"crypto"` / `"debit"` / `"apm"`), `amount_cents` (int), `fee_cents`
+(int), `created_ts` (int Unix seconds), and `finalized_ts: int | None`
+which is `None` until the transfer settles.
+
+Both `*_all` variants accept `max_pages=N` to bound iteration.
+
 ## Position fields
 
 `MarketPosition` and `EventPosition` use the standard `_dollars` / `_fp`
diff --git a/docs/types.md b/docs/types.md
index 02e6763..b044c58 100644
--- a/docs/types.md
+++ b/docs/types.md
@@ -44,12 +44,20 @@ Some fields are plain `int` cents, **not** `DollarDecimal`:
 | Field | Where | Unit |
 |---|---|---|
 | `Balance.balance` / `portfolio_value` | `client.portfolio.balance()` | cents |
+| `Deposit.amount_cents` / `fee_cents` | `client.portfolio.deposits()` | cents |
+| `Withdrawal.amount_cents` / `fee_cents` | `client.portfolio.withdrawals()` | cents |
 | `CreateOrderRequest.buy_max_cost` | `client.orders.create(..., buy_max_cost=500)` | cents — `500` means $5.00 |
 | `ApplySubaccountTransferRequest.amount_cents` | `client.subaccounts.transfer(...)` | cents |
 
 Passing a `Decimal` or `float` to one of these raises `ValueError` at
 construction — by design, so silent ×100 corruption can't happen.
 
+!!! warning "`Balance.balance` vs. `IndexedBalance.balance` — same name, different type"
+    `Balance.balance` is integer **cents**. `IndexedBalance.balance` (inside
+    `Balance.balance_breakdown`) is `DollarDecimal` (**dollars**) per the
+    spec. The dollars-form of the top-level total is exposed separately as
+    `Balance.balance_dollars` (required since v2.1.0).
+
 ## Nullable lists
 
 `NullableList[T]` coerces a server `null` into `[]`. Used on response fields
@@ -64,7 +72,8 @@ Every enum-like kwarg uses a `Literal` alias so your IDE auto-completes and
 
 | Alias | Values |
 |---|---|
-| `SideLiteral` | `"yes"`, `"no"` |
+| `SideLiteral` | `"yes"`, `"no"` (V1 orders) |
+| `BookSideLiteral` | `"bid"`, `"ask"` (V2 event-market orders) |
 | `ActionLiteral` | `"buy"`, `"sell"` |
 | `TimeInForceLiteral` | `"fill_or_kill"`, `"good_till_canceled"`, `"immediate_or_cancel"` |
 | `SelfTradePreventionTypeLiteral` | `"taker_at_cross"`, `"maker"` |
@@ -77,6 +86,9 @@ Every enum-like kwarg uses a `Literal` alias so your IDE auto-completes and
 | `SettlementStatusLiteral` | `"all"`, `"unsettled"`, `"settled"` |
 | `IncentiveProgramStatusLiteral` | `"all"`, `"active"`, `"upcoming"`, `"closed"`, … |
 | `IncentiveProgramTypeLiteral` | `"all"`, `"liquidity"`, `"volume"` |
+| `UserFilterLiteral` | `"self"` (server-side shorthand for the caller's user-id; see [Communications](resources/communications.md#filtering-shortcuts-v210)) |
+| `PaymentStatusLiteral` | `"pending"`, `"applied"`, `"failed"`, `"returned"` — used on `Deposit.status` and `Withdrawal.status` |
+| `PaymentTypeLiteral` | `"ach"`, `"wire"`, `"crypto"`, `"debit"`, `"apm"` — used on `Deposit.type` and `Withdrawal.type` |
 
 !!! tip "Markets pause; events don't"
     `MarketStatusLiteral` has a `"paused"` value that `EventStatusLiteral`