opencode-cmd-provider 1.5.0 → 1.5.2

package/CHANGELOG.md

@@ -1,5 +1,70 @@
 # Changelog
 
+## 1.5.2 - 2026-08-28
+
+Chore: catalog refresh to `command-code@1.36.0`, with a deals-coverage gate
+fix for name-colliding free variants.
+
+- Added `z-ai/glm-5.3-flash` (GLM-5.3 Flash) and `Qwen/Qwen3.8-Flash` with
+  deals records; removed `stealth/ox-alpha` (dropped from the upstream
+  catalog).
+- MiniMax free variants (`minimax/minimax-m3-free`,
+  `minimax/minimax-m2.7-free`) now append `(free)` to their picker display
+  name — upstream renamed them to share the paid display name — derived
+  data-driven from the zero-cost table, with updated deal allowance pricing
+  for `deepseek/deepseek-v4-flash-vision-exp`.
+- The deals-coverage gate now iterates every snapshot entry by id, so free
+  variants whose display names collide with paid siblings are actually
+  coverage-checked (a fixture that silently dropped them previously passed
+  with exit 0 and emitted a partial deals catalog); a shared
+  `scripts/snapshot-index.mjs` parser replaced the duplicated regex parsing.
+
+### Fixes
+
+- **Deals-coverage gate**: `check-deals-coverage.mjs` and the
+  `missingDealsModels` check in `refresh-deals.mjs` iterated a first-wins
+  name→id map, so MiniMax free variants — whose snapshot display names
+  collide with their paid siblings — were never actually checked. Both gates
+  now iterate by id, with the name fallback applying only to unambiguous
+  names; records that resolve to no snapshot id are skipped instead of
+  collapsing under `undefined`.
+- Catalog refresh to `command-code@1.33.0` added the MiniMax free variants;
+  the `1.36.0` refresh added GLM-5.3 Flash and Qwen 3.8 Flash, dropped Ox
+  Alpha, and extended `src/plugin/models.ts` with the zero-cost `(free)`
+  suffix. The extended pricing import now fits the print width again
+  (`format:check` passes).
+
+### Chores
+
+- New `tests/deals-coverage.test.ts` regression cases: removed free variants
+  now fail the gate; deals free flag ⇔ facts zero-cost table consistency;
+  negative suffix test (absent cost entry must not get `(free)`).
+- README documents the `(free)` suffix and its zero-cost derivation.
+
+## 1.5.1 - 2026-08-24
+
+Fix: provider parsers now synthesize the full reasoning and text lifecycle —
+`reasoning-start`/`reasoning-end` and `text-start`/`text-end` — for both
+OpenAI-style and Anthropic-style streams, so AI SDK consumers see balanced
+section boundaries instead of missing or duplicated end events.
+
+### Fixes
+
+- **OpenAI-style streams** now emit `reasoning-start` before the first
+  reasoning delta and `reasoning-end` before text, tool calls, or finish; the
+  same for `text-start`/`text-end`. Reasoning and text sections use stable ids
+  from the first chunk (instead of per-chunk ids), and unfinished sections are
+  closed when the stream finishes.
+- **Anthropic-style streams** now recognize `thinking` blocks and emit
+  `reasoning-start`/`reasoning-delta`/`reasoning-end` for them. `content_block_stop`
+  closes each block with the correct end event per its type (`text-end`,
+  `reasoning-end`, or `tool-input-end` + a single `tool-call`), replacing the
+  previous "emit both text-end and tool-input-end" pair that left consumers to
+  ignore the spurious one.
+- New `tests/stream.test.ts` cases cover the OpenAI reasoning→text ordering,
+  reasoning-only finishes, and the Anthropic thinking-block lifecycle;
+  `tests/provider-parity.test.ts` accepts `text-start` as the first text part.
+
 ## 1.5.0 - 2026-08-23
 
 Feature: the `[CMD] ` display-name prefix for auto-registered models is now

package/README.md

@@ -128,10 +128,15 @@ bundled. Core (models, auth, streaming) is unaffected.
 The plugin ships two bundled catalogs and auto-registers every model into
 OpenCode's config at startup, with the `[CMD]` display-name prefix (e.g.
 `[CMD] Claude Sonnet 5`) so they aren't confused with same-named models from
-other providers. Model availability changes when the package is updated. You
-can still declare your own `provider.commandcode` entry; your declarations
-always win and the snapshot fills in only what's missing (`whitelist`/
-`blacklist` on a declared entry filter the auto-registered models too).
+other providers. Free-tier variants that share an upstream display name with a
+paid model (e.g. the MiniMax M3 free tier) get a `(free)` suffix —
+`[CMD] MiniMax M3 (free)` — so the two are distinguishable in the model
+picker. The suffix is derived from the bundled pricing table: a model is
+labeled free only when the catalog carries an explicit zero-cost entry. Model
+availability changes when the package is updated. You can still declare your
+own `provider.commandcode` entry; your declarations always win and the
+snapshot fills in only what's missing (`whitelist`/`blacklist` on a declared
+entry filter the auto-registered models too).
 
 The `[CMD] ` display-name prefix is configurable through the declared
 provider entry's options:

package/dist/src/catalog/facts.d.ts

@@ -1,7 +1,7 @@
-export declare const FACTS_SOURCE_URL = "https://unpkg.com/command-code@1.32.1/dist/bundled/command-code-knowledge/reference/models.md";
-export declare const MODALITIES_SOURCE_URL = "https://unpkg.com/command-code@1.32.1/dist/cli.mjs";
-export declare const FACTS_PACKAGE_VERSION = "1.32.1";
-export declare const FACTS_LAST_REFRESHED = "2026-08-23";
+export declare const FACTS_SOURCE_URL = "https://unpkg.com/command-code@1.36.0/dist/bundled/command-code-knowledge/reference/models.md";
+export declare const MODALITIES_SOURCE_URL = "https://unpkg.com/command-code@1.36.0/dist/cli.mjs";
+export declare const FACTS_PACKAGE_VERSION = "1.36.0";
+export declare const FACTS_LAST_REFRESHED = "2026-08-27";
 export declare const MODEL_EFFORTS: Readonly<Record<string, readonly string[]>>;
 export declare const MODEL_COSTS: Readonly<Record<string, {
     input: number;

package/dist/src/catalog/facts.js

@@ -4,19 +4,20 @@
 // command-code npm package's bundled model catalog (models.md), plus input
 // modalities parsed from its CLI bundle (dist/cli.mjs). Regenerate
 // with `npm run refresh:snapshot`.
-export const FACTS_SOURCE_URL = "https://unpkg.com/command-code@1.32.1/dist/bundled/command-code-knowledge/reference/models.md";
-export const MODALITIES_SOURCE_URL = "https://unpkg.com/command-code@1.32.1/dist/cli.mjs";
-export const FACTS_PACKAGE_VERSION = "1.32.1";
-export const FACTS_LAST_REFRESHED = "2026-08-23";
+export const FACTS_SOURCE_URL = "https://unpkg.com/command-code@1.36.0/dist/bundled/command-code-knowledge/reference/models.md";
+export const MODALITIES_SOURCE_URL = "https://unpkg.com/command-code@1.36.0/dist/cli.mjs";
+export const FACTS_PACKAGE_VERSION = "1.36.0";
+export const FACTS_LAST_REFRESHED = "2026-08-27";
 export const MODEL_EFFORTS = {
     "deepseek/deepseek-v4-pro": ["high", "max"],
     "deepseek/deepseek-v4-flash": ["high", "max"],
     "deepseek/deepseek-v4-flash-vision-exp": ["high", "max"],
+    "z-ai/glm-5.3-flash": ["low", "high", "max"],
     "zai-org/GLM-5.3": ["low", "high", "max"],
     "zai-org/GLM-5.2": ["high", "max"],
     "Qwen/Qwen3.8-Max": ["low", "medium", "xhigh"],
     "Qwen/Qwen3.8-27B": ["low", "medium", "xhigh"],
-    "stealth/ox-alpha": ["low", "high", "max"],
+    "Qwen/Qwen3.8-Flash": ["low", "medium", "xhigh"],
     "claude-sonnet-5": ["low", "medium", "high", "xhigh", "max"],
     "claude-sonnet-4-6": ["low", "medium", "high", "xhigh", "max"],
     "claude-fable-5": ["low", "medium", "high", "xhigh", "max"],
@@ -42,12 +43,13 @@ export const MODEL_EFFORTS = {
 export const MODEL_COSTS = {
     "deepseek/deepseek-v4-pro": { input: 0.66, output: 1.98, cacheRead: 0.022, cacheWrite: 0 },
     "deepseek/deepseek-v4-flash": { input: 0.22, output: 0.66, cacheRead: 0.007, cacheWrite: 0 },
-    "deepseek/deepseek-v4-flash-vision-exp": { input: 0.22, output: 0.66, cacheRead: 0.01, cacheWrite: 0 },
+    "deepseek/deepseek-v4-flash-vision-exp": { input: 0.22, output: 0.66, cacheRead: 0.007, cacheWrite: 0 },
     "moonshotai/Kimi-K3": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 0 },
     "moonshotai/Kimi-K2.7-Code": { input: 0.95, output: 4, cacheRead: 0.19, cacheWrite: 0 },
     "moonshotai/Kimi-K2.7-Code-Highspeed": { input: 1.9, output: 8, cacheRead: 0.38, cacheWrite: 0 },
     "moonshotai/Kimi-K2.6": { input: 0.95, output: 4, cacheRead: 0.16, cacheWrite: 0 },
     "moonshotai/Kimi-K2.5": { input: 0.6, output: 3, cacheRead: 0.1, cacheWrite: 0 },
+    "z-ai/glm-5.3-flash": { input: 0.15, output: 0.5, cacheRead: 0.03, cacheWrite: 0 },
     "zai-org/GLM-5.3": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 },
     "zai-org/GLM-5.2": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 },
     "zai-org/GLM-5.2-Fast": { input: 3, output: 10.25, cacheRead: 0.5, cacheWrite: 0 },
@@ -55,11 +57,14 @@ export const MODEL_COSTS = {
     "zai-org/GLM-5": { input: 1, output: 3.2, cacheRead: 0.2, cacheWrite: 0 },
     "MiniMaxAI/MiniMax-M3": { input: 0.3, output: 1.2, cacheRead: 0.06, cacheWrite: 0 },
     "MiniMaxAI/MiniMax-M2.7": { input: 0.3, output: 1.2, cacheRead: 0.06, cacheWrite: 0 },
+    "minimax/minimax-m3-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
+    "minimax/minimax-m2.7-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
     "MiniMaxAI/MiniMax-M2.5": { input: 0.3, output: 1.2, cacheRead: 0.03, cacheWrite: 0 },
     "xiaomi/mimo-v2.5-pro": { input: 0.435, output: 0.87, cacheRead: 0.0036, cacheWrite: 0 },
     "xiaomi/mimo-v2.5": { input: 0.14, output: 0.28, cacheRead: 0.0028, cacheWrite: 0 },
     "Qwen/Qwen3.8-Max": { input: 2, output: 6, cacheRead: 0.25, cacheWrite: 2.5 },
     "Qwen/Qwen3.8-27B": { input: 0.4, output: 3, cacheRead: 0.04, cacheWrite: 0 },
+    "Qwen/Qwen3.8-Flash": { input: 0.16, output: 0.47, cacheRead: 0.016, cacheWrite: 0 },
     "Qwen/Qwen3.7-Max": { input: 2.5, output: 7.5, cacheRead: 0.5, cacheWrite: 3.13 },
     "Qwen/Qwen3.7-Plus": { input: 0.4, output: 1.6, cacheRead: 0.08, cacheWrite: 0.5 },
     "Qwen/Qwen3.7-Flash": { input: 0.03, output: 0.13, cacheRead: 0.006, cacheWrite: 0.038 },
@@ -72,7 +77,6 @@ export const MODEL_COSTS = {
     "thinkingmachines/inkling": { input: 1, output: 4.05, cacheRead: 0.17, cacheWrite: 0 },
     "thinkingmachines/inkling-small": { input: 0.5, output: 1.2, cacheRead: 0.1, cacheWrite: 0 },
     "poolside/laguna-s-2.1-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
-    "stealth/ox-alpha": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },
     "claude-sonnet-5": { input: 2, output: 10, cacheRead: 0.2, cacheWrite: 2.5 },
     "claude-sonnet-4-6": { input: 3, output: 15, cacheRead: 0.3, cacheWrite: 3.75 },
     "claude-fable-5": { input: 10, output: 50, cacheRead: 1, cacheWrite: 12.5 },
@@ -105,6 +109,7 @@ export const MODEL_INPUT_MODALITIES = {
     "Qwen/Qwen3.7-Flash": ["text", "image"],
     "Qwen/Qwen3.7-Plus": ["text", "image"],
     "Qwen/Qwen3.8-27B": ["text", "image"],
+    "Qwen/Qwen3.8-Flash": ["text", "image"],
     "Qwen/Qwen3.8-Max": ["text", "image"],
     "claude-fable-5": ["text", "image"],
     "claude-haiku-4-5-20251001": ["text", "image"],
@@ -129,16 +134,17 @@ export const MODEL_INPUT_MODALITIES = {
     "meta/muse-spark-1.1": ["text", "image"],
     "meta/muse-spark-1.2": ["text", "image"],
     "meta/muse-spark-1.2-contributor": ["text", "image"],
+    "minimax/minimax-m3-free": ["text", "image"],
     "moonshotai/Kimi-K2.5": ["text", "image"],
     "moonshotai/Kimi-K2.6": ["text", "image"],
     "moonshotai/Kimi-K2.7-Code": ["text", "image"],
     "moonshotai/Kimi-K2.7-Code-Highspeed": ["text", "image"],
     "moonshotai/Kimi-K3": ["text", "image"],
     "sakana/fugu-ultra": ["text", "image"],
-    "stealth/ox-alpha": ["text", "image"],
     "stepfun/Step-3.7-Flash": ["text", "image"],
     "thinkingmachines/inkling": ["text", "image"],
     "thinkingmachines/inkling-small": ["text", "image"],
     "xai/grok-4.5": ["text", "image"],
     "xiaomi/mimo-v2.5": ["text", "image"],
+    "z-ai/glm-5.3-flash": ["text", "image"],
 };

package/dist/src/catalog/snapshot.js

@@ -26,6 +26,7 @@ export const MODEL_SNAPSHOT = [
     { id: "moonshotai/Kimi-K2.7-Code-Highspeed", name: "Kimi K2.7 Code HighSpeed", contextLength: 262000 },
     { id: "moonshotai/Kimi-K2.6", name: "Kimi K2.6", contextLength: 256000 },
     { id: "moonshotai/Kimi-K2.5", name: "Kimi K2.5", contextLength: 256000 },
+    { id: "z-ai/glm-5.3-flash", name: "GLM-5.3 Flash", contextLength: 1048576 },
     { id: "zai-org/GLM-5.3", name: "GLM-5.3", contextLength: 1000000 },
     { id: "zai-org/GLM-5.2", name: "GLM-5.2", contextLength: 1000000 },
     { id: "zai-org/GLM-5.2-Fast", name: "GLM-5.2 Fast", contextLength: 1000000 },
@@ -33,11 +34,14 @@ export const MODEL_SNAPSHOT = [
     { id: "zai-org/GLM-5", name: "GLM-5", contextLength: 200000 },
     { id: "MiniMaxAI/MiniMax-M3", name: "MiniMax M3", contextLength: 1000000 },
     { id: "MiniMaxAI/MiniMax-M2.7", name: "MiniMax M2.7", contextLength: 200000 },
+    { id: "minimax/minimax-m3-free", name: "MiniMax M3", contextLength: 1000000 },
+    { id: "minimax/minimax-m2.7-free", name: "MiniMax M2.7", contextLength: 197000 },
     { id: "MiniMaxAI/MiniMax-M2.5", name: "MiniMax M2.5", contextLength: 200000 },
     { id: "xiaomi/mimo-v2.5-pro", name: "MiMo V2.5 Pro", contextLength: 1000000 },
     { id: "xiaomi/mimo-v2.5", name: "MiMo V2.5", contextLength: 1000000 },
     { id: "Qwen/Qwen3.8-Max", name: "Qwen 3.8 Max", contextLength: 1000000 },
     { id: "Qwen/Qwen3.8-27B", name: "Qwen 3.8 27B", contextLength: 262144 },
+    { id: "Qwen/Qwen3.8-Flash", name: "Qwen 3.8 Flash", contextLength: 1000000 },
     { id: "Qwen/Qwen3.7-Max", name: "Qwen 3.7 Max", contextLength: 1000000 },
     { id: "Qwen/Qwen3.7-Plus", name: "Qwen 3.7 Plus", contextLength: 1000000 },
     { id: "Qwen/Qwen3.7-Flash", name: "Qwen 3.7 Flash", contextLength: 1000000 },
@@ -55,7 +59,6 @@ export const MODEL_SNAPSHOT = [
     { id: "nvidia/nemotron-3-ultra-550b-a55b", name: "Nemotron 3 Ultra", contextLength: 1000000 },
     { id: "thinkingmachines/inkling", name: "Inkling", contextLength: 256000 },
     { id: "thinkingmachines/inkling-small", name: "Inkling Small", contextLength: 1000000 },
-    { id: "stealth/ox-alpha", name: "Ox Alpha", contextLength: 1048576 },
     { id: "poolside/laguna-s-2.1-free", name: "Laguna S 2.1", contextLength: 256000 },
     { id: "meta/muse-spark-1.1", name: "Muse Spark 1.1", contextLength: 1048576 },
     { id: "meta/muse-spark-1.2", name: "Muse Spark 1.2", contextLength: 1048576 },

package/dist/src/deals/catalog.d.ts

@@ -51,5 +51,5 @@ export interface ModelDeals {
 export declare const MODEL_DEALS: Readonly<Record<string, ModelDeals>>;
 export declare const PLAN_CATALOG: Readonly<Record<PlanId, PlanInfo>>;
 export declare const DEAL_SOURCE_URL = "https://commandcode.ai/docs/resources/pricing-limits";
-export declare const DEAL_LAST_REFRESHED = "2026-08-23";
+export declare const DEAL_LAST_REFRESHED = "2026-08-27";
 export declare const DEAL_PACKAGE_VERSION = "docs";

package/dist/src/deals/catalog.js

@@ -13,7 +13,7 @@ export const MODEL_DEALS = {
     "claude-sonnet-4-6": { tier: "premium", benchmark: { "intelligence": 48.4 }, free: false, allowance: { "pro": 20 } },
     "claude-sonnet-5": { tier: "premium", benchmark: { "intelligence": 55.3, "tokPerSec": 81.9 }, free: false, allowance: { "pro": 20 } },
     "deepseek/deepseek-v4-flash": { tier: "opensource", benchmark: { "intelligence": 51.8, "tokPerSec": 114.6 }, peakOffPeak: { "peak": { "input": 0.44, "output": 1.32, "cacheRead": 0.014, "cacheWrite": 0 }, "offPeak": { "input": 0.22, "output": 0.66, "cacheRead": 0.007, "cacheWrite": 0 }, "windows": "01–04 & 06–10 UTC" }, free: false, allowance: { "goat": 60, "pro": 70 } },
-    "deepseek/deepseek-v4-flash-vision-exp": { tier: "opensource", peakOffPeak: { "peak": { "input": 0.44, "output": 1.32, "cacheRead": 0.02, "cacheWrite": 0 }, "offPeak": { "input": 0.22, "output": 0.66, "cacheRead": 0.01, "cacheWrite": 0 }, "windows": "01–04 & 06–10 UTC" }, free: false, allowance: { "goat": 20, "pro": 30 } },
+    "deepseek/deepseek-v4-flash-vision-exp": { tier: "opensource", peakOffPeak: { "peak": { "input": 0.44, "output": 1.32, "cacheRead": 0.014, "cacheWrite": 0 }, "offPeak": { "input": 0.22, "output": 0.66, "cacheRead": 0.007, "cacheWrite": 0 }, "windows": "01–04 & 06–10 UTC" }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "deepseek/deepseek-v4-pro": { tier: "opensource", benchmark: { "intelligence": 53.2, "tokPerSec": 75.3 }, peakOffPeak: { "peak": { "input": 1.32, "output": 3.96, "cacheRead": 0.044, "cacheWrite": 0 }, "offPeak": { "input": 0.66, "output": 1.98, "cacheRead": 0.022, "cacheWrite": 0 }, "windows": "01–04 & 06–10 UTC" }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "google/gemini-3.1-flash-lite": { tier: "premium", benchmark: { "intelligence": 25.6 }, free: false, allowance: { "pro": 20 } },
     "google/gemini-3.5-flash": { tier: "premium", benchmark: { "intelligence": 52 }, free: false, allowance: { "pro": 20 } },
@@ -30,6 +30,8 @@ export const MODEL_DEALS = {
     "meta/muse-spark-1.1": { tier: "premium", benchmark: { "intelligence": 53.2 }, free: false, allowance: { "pro": 20 } },
     "meta/muse-spark-1.2": { tier: "opensource", benchmark: { "intelligence": 56.8 }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "meta/muse-spark-1.2-contributor": { tier: "opensource", benchmark: { "intelligence": 56.8 }, free: false, allowance: { "goat": 20, "pro": 30 } },
+    "minimax/minimax-m2.7-free": { tier: "opensource", free: true },
+    "minimax/minimax-m3-free": { tier: "opensource", free: true },
     "MiniMaxAI/MiniMax-M2.5": { tier: "opensource", benchmark: { "intelligence": 34.5 }, free: false },
     "MiniMaxAI/MiniMax-M2.7": { tier: "opensource", benchmark: { "intelligence": 38.9 }, free: false },
     "MiniMaxAI/MiniMax-M3": { tier: "opensource", discount: { "pct": 50 }, was: { "input": 0.6, "output": 2.4, "cacheRead": 0.12 }, now: { "input": 0.3, "output": 1.2, "cacheRead": 0.06 }, benchmark: { "intelligence": 45.4, "tokPerSec": 104.4 }, free: false, allowance: { "goat": 47, "pro": 57 } },
@@ -46,9 +48,9 @@ export const MODEL_DEALS = {
     "Qwen/Qwen3.7-Max": { tier: "opensource", benchmark: { "intelligence": 46.7 }, free: false, allowance: { "goat": 33, "pro": 43 } },
     "Qwen/Qwen3.7-Plus": { tier: "opensource", benchmark: { "intelligence": 39.4, "tokPerSec": 55.6 }, overContext: { "input": 1.2, "output": 4.8, "cacheRead": 0.24, "cacheWrite": 1.5 }, free: false, allowance: { "goat": 33, "pro": 43 } },
     "Qwen/Qwen3.8-27B": { tier: "opensource", benchmark: { "intelligence": 52 }, free: false, allowance: { "goat": 70, "pro": 80 } },
+    "Qwen/Qwen3.8-Flash": { tier: "opensource", free: false, allowance: { "goat": 20, "pro": 30 } },
     "Qwen/Qwen3.8-Max": { tier: "opensource", benchmark: { "intelligence": 58.1, "tokPerSec": 46.7 }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "sakana/fugu-ultra": { tier: "premium", free: false },
-    "stealth/ox-alpha": { tier: "opensource", free: true },
     "stepfun/Step-3.5-Flash": { tier: "opensource", benchmark: { "intelligence": 26.5 }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "stepfun/Step-3.7-Flash": { tier: "opensource", benchmark: { "intelligence": 30.9, "tokPerSec": 120.1 }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "tencent/hy3-paid": { tier: "opensource", benchmark: { "intelligence": 42.2, "tokPerSec": 74.8 }, free: false, allowance: { "goat": 70, "pro": 80 } },
@@ -58,6 +60,7 @@ export const MODEL_DEALS = {
     "xai/grok-4.6": { tier: "opensource", benchmark: { "intelligence": 60.9, "tokPerSec": 66 }, overContext: { "input": 4, "output": 12, "cacheRead": 1, "cacheWrite": 0 }, free: false, allowance: { "goat": 20, "pro": 30 } },
     "xiaomi/mimo-v2.5": { tier: "opensource", discount: { "pct": 98 }, was: { "input": 0.8, "output": 4, "cacheRead": 0.16 }, now: { "input": 0.14, "output": 0.28, "cacheRead": 0.0028 }, benchmark: { "intelligence": 38, "tokPerSec": 61.8 }, free: false, allowance: { "goat": 30, "pro": 40 } },
     "xiaomi/mimo-v2.5-pro": { tier: "opensource", discount: { "pct": 99 }, was: { "input": 2, "output": 6, "cacheRead": 0.4 }, now: { "input": 0.435, "output": 0.87, "cacheRead": 0.0036 }, benchmark: { "intelligence": 42.9, "tokPerSec": 58.1 }, free: false, allowance: { "goat": 20, "pro": 30 } },
+    "z-ai/glm-5.3-flash": { tier: "opensource", free: false, allowance: { "goat": 40, "pro": 50 } },
     "zai-org/GLM-5": { tier: "opensource", benchmark: { "intelligence": 40.6 }, free: false },
     "zai-org/GLM-5.1": { tier: "opensource", benchmark: { "intelligence": 41, "tokPerSec": 28.2 }, free: false },
     "zai-org/GLM-5.2": { tier: "opensource", benchmark: { "intelligence": 52.6, "tokPerSec": 69 }, free: false, allowance: { "goat": 70, "pro": 80 } },
@@ -74,5 +77,5 @@ export const PLAN_CATALOG = {
     provider: { price: 15, credits: 0, window5h: 0, windowWeek: 0, display: "Provider" },
 };
 export const DEAL_SOURCE_URL = "https://commandcode.ai/docs/resources/pricing-limits";
-export const DEAL_LAST_REFRESHED = "2026-08-23";
+export const DEAL_LAST_REFRESHED = "2026-08-27";
 export const DEAL_PACKAGE_VERSION = "docs";

package/dist/src/plugin/models.js

@@ -1,4 +1,4 @@
-import { MODEL_COSTS, ZERO_MODEL_COST } from "../provider/pricing.js";
+import { MODEL_COSTS, ZERO_MODEL_COST, isFreeModelCost, } from "../provider/pricing.js";
 import { reasoningVariantsForModel, isReasoningModel } from "../provider/reasoning.js";
 import { inputModalitiesForModel } from "../provider/modalities.js";
 const DEFAULT_MAX_OUTPUT_TOKENS = 65_536;
@@ -95,8 +95,14 @@ export function augmentConfigCommandCodeModels(config) {
 function configModelFor(model, prefix = DEFAULT_DISPLAY_PREFIX) {
     const variants = reasoningVariantsForModel(model.id);
     const costs = MODEL_COSTS[model.id] ?? ZERO_MODEL_COST;
+    // Upstream names the paid and free MiniMax variants identically
+    // (`MiniMax M3` / `MiniMax M2.7`); append `(free)` to the picker entry so
+    // users can tell them apart. Data-driven from the zero cost table — a model
+    // is only "free" when the catalog has an actual zero-cost entry (an absent
+    // entry falls back to ZERO_MODEL_COST and is NOT free).
+    const freeSuffix = MODEL_COSTS[model.id] !== undefined && isFreeModelCost(costs) ? " (free)" : "";
     return {
-        name: `${prefix}${model.name}`,
+        name: `${prefix}${model.name}${freeSuffix}`,
         limit: {
             context: model.contextLength,
             output: Math.min(model.contextLength, DEFAULT_MAX_OUTPUT_TOKENS),

package/dist/src/provider/pricing.d.ts

@@ -9,3 +9,10 @@ export interface CommandCodeModelCostRates {
 export interface CommandCodeModelCost extends CommandCodeModelCostRates {
 }
 export declare const ZERO_MODEL_COST: CommandCodeModelCost;
+/**
+ * True when every rate is zero — the model is served free (e.g. the MiniMax
+ * M3 / M2.7 free variants). Used to disambiguate display names: the upstream
+ * catalog names the paid and free variants identically (`MiniMax M3`), so the
+ * auto-registered picker entry appends `(free)` when this matches.
+ */
+export declare function isFreeModelCost(cost: CommandCodeModelCost | undefined): boolean;

package/dist/src/provider/pricing.js

@@ -8,3 +8,16 @@ export const ZERO_MODEL_COST = {
     cacheRead: 0,
     cacheWrite: 0,
 };
+/**
+ * True when every rate is zero — the model is served free (e.g. the MiniMax
+ * M3 / M2.7 free variants). Used to disambiguate display names: the upstream
+ * catalog names the paid and free variants identically (`MiniMax M3`), so the
+ * auto-registered picker entry appends `(free)` when this matches.
+ */
+export function isFreeModelCost(cost) {
+    return (cost !== undefined &&
+        cost.input === 0 &&
+        cost.output === 0 &&
+        cost.cacheRead === 0 &&
+        cost.cacheWrite === 0);
+}

package/dist/src/provider/reasoning.js

@@ -29,12 +29,12 @@ export const REASONING_MODELS = new Set([
     "meta/muse-spark-1.1",
     "meta/muse-spark-1.2",
     "meta/muse-spark-1.2-contributor",
+    "minimax/minimax-m3-free",
     "moonshotai/Kimi-K2.7-Code",
     "moonshotai/Kimi-K2.7-Code-Highspeed",
     "moonshotai/Kimi-K3",
     "nvidia/nemotron-3-ultra-550b-a55b",
     "poolside/laguna-s-2.1-free",
-    "stealth/ox-alpha",
     "stepfun/Step-3.7-Flash",
     "tencent/hy3-paid",
     "thinkingmachines/inkling",

package/dist/src/provider/stream.js

@@ -338,6 +338,10 @@ export function anthropicEventToStreamPart(event) {
             const id = `text-${index}`;
             return [{ type: "text-start", id }];
         }
+        if (blockType === "thinking") {
+            const id = stringValue(block?.id) ?? `thinking-${index}`;
+            return [{ type: "reasoning-start", id }];
+        }
         if (blockType === "tool_use") {
             const id = stringValue(block?.id) ?? `tool-${index}`;
             const name = stringValue(block?.name) ?? "";
@@ -345,16 +349,32 @@ export function anthropicEventToStreamPart(event) {
         }
         return [];
     }
+    if (type === "content_block_delta") {
+        const delta = asRecord(event.delta);
+        const index = numberValue(event.index) ?? 0;
+        const text = stringValue(delta?.text);
+        if (typeof text === "string" && text.length > 0) {
+            const id = `text-${index}`;
+            return [{ type: "text-delta", id, delta: text }];
+        }
+        const thinking = stringValue(delta?.thinking);
+        if (typeof thinking === "string" && thinking.length > 0) {
+            const id = `thinking-${index}`;
+            return [{ type: "reasoning-delta", id, delta: thinking }];
+        }
+        // Tool input delta: { type: "input_json_delta", partial_json: "..." }
+        const partial = stringValue(delta?.partial_json);
+        if (typeof partial === "string" && partial.length > 0) {
+            const id = `tool-${index}`;
+            // Emit as tool-input-delta; caller may have started tool
+            return [{ type: "tool-input-delta", id, delta: partial }];
+        }
+        return [];
+    }
     if (type === "content_block_stop") {
         // Stateless codec: the STOP event carries only `index`, not the block type.
-        // Anthropic streams either `text` or `tool_use` blocks; the AI SDK expects
-        // `text-end` for the former and `tool-input-end` for the latter. Emitting
-        // both is noisy (previous emitted two parts) but guarantees the right end
-        // is seen regardless of block type, and the consumer ignores the spurious
-        // one per AI SDK spec. Emit a single text-end here would break tool_use
-        // streams that rely on tool-input-end, so we keep the conservative pair
-        // with an explicit note referencing provider doc streaming (message_delta
-        // carries the final usage/finish; per-block ends are AI SDK concerns).
+        // Anthropic streams either `text`, `thinking` or `tool_use` blocks; the AI SDK expects
+        // `text-end` for text, `reasoning-end` for thinking, and `tool-input-end` for tool_use.
         const index = numberValue(event.index) ?? 0;
         const idText = `text-${index}`;
         const idTool = `tool-${index}`;
@@ -395,6 +415,26 @@ export function createOpenAIStreamParser() {
     // finish_reason here so the usage-only chunk's finish keeps the real reason
     // (e.g. "length") instead of defaulting to "stop".
     let lastFinishReason;
+    let reasoningStarted = false;
+    let reasoningEnded = false;
+    let reasoningId;
+    let textStarted = false;
+    let textEnded = false;
+    let textId;
+    function closeReasoning() {
+        if (reasoningStarted && !reasoningEnded && reasoningId) {
+            reasoningEnded = true;
+            return [{ type: "reasoning-end", id: reasoningId }];
+        }
+        return [];
+    }
+    function closeText() {
+        if (textStarted && !textEnded && textId) {
+            textEnded = true;
+            return [{ type: "text-end", id: textId }];
+        }
+        return [];
+    }
     return (event) => {
         if (!isRecord(event))
             return [];
@@ -405,17 +445,42 @@ export function createOpenAIStreamParser() {
         const parts = [];
         const choice = firstChoice(event);
         const delta = choice ? deltaFromChoice(choice) : undefined;
+        const eventId = stringValue(event.id);
         if (delta) {
-            const content = stringValue(delta.content);
-            if (typeof content === "string" && content.length > 0) {
-                parts.push(textDeltaPart(event.id, content));
-            }
             const reasoning = stringValue(delta.reasoning) ?? stringValue(delta.reasoning_content);
             if (typeof reasoning === "string" && reasoning.length > 0) {
-                parts.push(reasoningDeltaPart(event.id, reasoning));
+                if (!reasoningStarted || reasoningEnded) {
+                    if (textStarted && !textEnded) {
+                        parts.push(...closeText());
+                    }
+                    reasoningStarted = true;
+                    reasoningEnded = false;
+                    reasoningId = eventId || "reasoning-0";
+                    parts.push({ type: "reasoning-start", id: reasoningId });
+                }
+                parts.push(reasoningDeltaPart(reasoningId ?? eventId, reasoning));
+            }
+            const content = stringValue(delta.content);
+            if (typeof content === "string" && content.length > 0) {
+                if (reasoningStarted && !reasoningEnded) {
+                    parts.push(...closeReasoning());
+                }
+                if (!textStarted || textEnded) {
+                    textStarted = true;
+                    textEnded = false;
+                    textId = eventId || "text-0";
+                    parts.push({ type: "text-start", id: textId });
+                }
+                parts.push(textDeltaPart(textId ?? eventId, content));
             }
             const toolCalls = delta.tool_calls ?? delta.toolCalls;
             if (Array.isArray(toolCalls)) {
+                if (reasoningStarted && !reasoningEnded) {
+                    parts.push(...closeReasoning());
+                }
+                if (textStarted && !textEnded) {
+                    parts.push(...closeText());
+                }
                 for (const tc of toolCalls) {
                     if (!isRecord(tc))
                         continue;
@@ -475,6 +540,12 @@ export function createOpenAIStreamParser() {
         if (finishReason)
             lastFinishReason = finishReason;
         if (lastFinishReason || hasUsage) {
+            if (reasoningStarted && !reasoningEnded) {
+                parts.push(...closeReasoning());
+            }
+            if (textStarted && !textEnded) {
+                parts.push(...closeText());
+            }
             // Flush any tool call whose terminal args chunk never arrived.
             for (const [index, buffer] of toolBuffers) {
                 if (buffer.started && !buffer.emitted) {
@@ -498,6 +569,7 @@ export function createOpenAIStreamParser() {
 }
 export function createAnthropicStreamParser() {
     const toolBlocks = new Map();
+    const blockTypes = new Map();
     return (event) => {
         if (!isRecord(event))
             return [];
@@ -507,13 +579,21 @@ export function createAnthropicStreamParser() {
             const index = numberValue(event.index) ?? 0;
             const blockType = stringValue(block?.type);
             if (blockType === "tool_use") {
+                blockTypes.set(index, "tool_use");
                 const id = stringValue(block?.id) ?? `tool-${index}`;
                 const name = stringValue(block?.name) ?? "";
                 toolBlocks.set(index, { id, name, input: "", started: true, emitted: false });
                 return [{ type: "tool-input-start", id, toolName: name }];
             }
-            if (blockType === "text")
+            if (blockType === "thinking") {
+                blockTypes.set(index, "thinking");
+                const id = stringValue(block?.id) ?? `thinking-${index}`;
+                return [{ type: "reasoning-start", id }];
+            }
+            if (blockType === "text") {
+                blockTypes.set(index, "text");
                 return [{ type: "text-start", id: `text-${index}` }];
+            }
             return [];
         }
         if (type === "content_block_delta") {
@@ -523,6 +603,10 @@ export function createAnthropicStreamParser() {
             if (typeof text === "string" && text.length > 0) {
                 return [{ type: "text-delta", id: `text-${index}`, delta: text }];
             }
+            const thinking = stringValue(delta?.thinking);
+            if (typeof thinking === "string" && thinking.length > 0) {
+                return [{ type: "reasoning-delta", id: `thinking-${index}`, delta: thinking }];
+            }
             const partial = stringValue(delta?.partial_json);
             if (typeof partial === "string" && partial.length > 0) {
                 const block = toolBlocks.get(index);
@@ -557,6 +641,28 @@ export function createAnthropicStreamParser() {
         }
         if (type === "content_block_stop") {
             const index = numberValue(event.index) ?? 0;
+            const bType = blockTypes.get(index);
+            blockTypes.delete(index);
+            if (bType === "tool_use") {
+                const block = toolBlocks.get(index);
+                if (block) {
+                    toolBlocks.delete(index);
+                    if (block.emitted)
+                        return [];
+                    return [
+                        { type: "tool-input-end", id: block.id },
+                        {
+                            type: "tool-call",
+                            toolCallId: block.id,
+                            toolName: block.name,
+                            input: block.input,
+                        },
+                    ];
+                }
+            }
+            else if (bType === "thinking") {
+                return [{ type: "reasoning-end", id: `thinking-${index}` }];
+            }
             const block = toolBlocks.get(index);
             if (block) {
                 toolBlocks.delete(index);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-cmd-provider",
-  "version": "1.5.0",
+  "version": "1.5.2",
   "description": "Command Code provider + plugin for opencode",
   "type": "module",
   "main": "./dist/index.js",