@lmctl-ai/lmctl 0.1.341 → 0.2.2

package/CHANGELOG.md

@@ -4,6 +4,137 @@ All notable public-preview changes for `@lmctl-ai/lmctl` are recorded here.
 
 ## Unreleased
 
+- **Shared output formatting via `@lmctl-ai/lmformat` (behavior).** All
+  human-readable CLI output now uses the lmformat library, vendored at
+  `vendor/lmctl-ai-lmformat-0.7.0.tgz` until the package is published:
+  - `status` and `ls` tables render with auto-sized columns and
+    right-aligned numeric columns (replacing hand-computed padding).
+  - Token counts are compacted everywhere (`960K in / 85K out (cached:
+    15.8M)` in `cost`, and in `status` single-member view).
+  - Durations unify on the two-unit style (`2h 5m`, `3m 12s`), ages carry
+    an `ago` suffix (`10m ago`), and week units kick in past 7 days.
+  - `running` and `ratelimit` render sectioned grids: standalone section
+    headings, a blank line between sections, one measured width set shared
+    across sections, column headers printed once above the first section,
+    bare cell values (no `model=`/`used`/`resets` in-cell labels), and the
+    open-ended timing column last.
+  - `ratelimit` reset countdowns are two-unit precise (`in 2d 4h`,
+    `13h ago`) and section headings use local `hh:mm:ss` — absolute ISO
+    timestamps moved to `--json`.
+  - **`ls --format table` MODIFIED is now relative (`3d ago`)** instead of
+    an absolute ISO timestamp; use `--format json` for exact timestamps.
+  - `--json` output, stderr narration, and stored timestamps are unchanged.
+
+## 0.2.1 — 2026-09-24
+
+- **`lmctl batch --bundle` (feature).** Batch can now send all prompts through
+  ONE provider process instead of respawning per paragraph, and holds the
+  member's in-flight lease for the whole bundle — an interloper chat can no
+  longer steal the session between prompts and abort the run. Supported for
+  every bundle-capable provider: the ACP five (`gemini`, `copilot`, `opencode`,
+  `kimi`, `hermes`), stream-json (`claude`, `qwen`), `codex` (app-server), and
+  `dsh` (sidecar). Per-turn bookkeeping (chat pointer, tracked_invocation under
+  the batch root, team_chat_log rows with `message_type=batch`) is unchanged.
+  Bundle resume is strict everywhere: ACP `session/load` must echo the exact
+  teamfile session id, stream-json verifies the on-disk session file's header
+  (at open and between prompts), codex disables its thread/start fallback, and
+  dsh requires the session file to exist. Cancellation keeps the bundle alive
+  where the transport has a cancel verb (ACP `session/cancel`, codex
+  `turn/interrupt`) and is terminal where it does not (stream-json, dsh). A
+  provider-process exit mid-bundle aborts with the completed prompt count;
+  lease loss between prompts aborts with a clear message. Refused (exit 2) for
+  agy/pi/lmplayer, and for providers with an `agent=` override in the user
+  config.
+
+- **lmprice model-catalog integration (feature, non-breaking).** The lmprice
+  published dataset (`model_list.json` + `aliases.json`) is now lmctl's model
+  source of truth:
+  - `lint`/`seed` resolve `model=` for `claude`, `gemini`, `qwen`, and
+    `hermes` against the catalog instead of static tested-model lists (which
+    are removed). Catalog misses warn
+    (`model "x" not found in lmprice catalog … provider may still accept it`)
+    and flow through the standard warnings array, so `lint --strict`
+    promotes them as before. Deliberately `unresolved` aliases are not
+    warnings.
+  - **"If we can lint, then we can cost" (operator invariant,
+    2026-09-25).** A recognized-but-unpriceable model (`known-unpriced`)
+    is now a lint/seed warning
+    (`model "x" is recognized in the lmprice catalog but has no price
+    information — cost will be unavailable`, `--strict`-promotable) instead
+    of a silent accept that cost then couldn't price. `unresolved` stays
+    silent. The lmprice data side is tasked with rejecting such
+    null-terminal bindings at publish time.
+  - `cost` resolves the session model provider-aware through the same
+    catalog: aliases.json pin when present, route preference
+    (first-party > litellm > openrouter > manual) otherwise. New JSON fields:
+    `priceResolution`, `priceRoute`, `priceProvenance`, `priceEstimated`,
+    `costUnavailableReason` (existing fields unchanged). Manual/estimated
+    rows are marked; unresolved/known-unpriced report cost-unavailable with
+    a reason instead of a guessed suffix match.
+  - Price-data loader now manages both published files with independent
+    etags, a 6h TTL (`LMCTL_PRICE_TTL`, `0` forces), validated atomic cache
+    promotion, and `LMCTL_PRICE_DATA=<dir>` snapshot pinning that disables
+    network refresh. The build bundles both files
+    (`data/price-aliases.json` is new); the pack ships both.
+  - **Best-effort contract (operator, 2026-09-25).** Model validation is
+    advisory and not claimed to be 100% current — it can lag provider
+    changes in either direction, so lint/seed warn but never error on
+    unrecognized or out-of-date models, and cost reports n/a with a reason
+    instead of failing. Retired-model example text updated
+    (`github-copilot/gpt-5.4` → `github-copilot/gpt-5.5`).
+- **New `ratelimit` command (stateless, real-time).** `lmctl ratelimit
+  [provider ...]` queries provider subscription rate limits live via the
+  external `lmratelimit` readers (binary resolved from `LMCTL_RATELIMIT_BIN`,
+  then `~/repos/tools/lmratelimit/bin/lmratelimit`, then `PATH`; `--timeout
+  <seconds>` sets the per-provider timeout, default 15) and renders
+  per-provider status, per-bucket used/remaining percentages with EXHAUSTED
+  and limit flags, plus credits/spend-control when present. `--json` emits
+  the full raw `{"readings": [...]}` payload. Nothing is read from or
+  written to any database — no collection, no storage, no forecast.
+
+## 0.2.0 — 2026-09-24
+
+- **Breaking CLI consolidation.** Eleven verbs are removed outright and their
+  workflows are absorbed into the remaining surface:
+  - `refresh` → removed; to re-seed an existing member's prompt, delete its
+    `sessionid=` field from the teamfile and re-run `lmctl seed` (plain seed
+    touches only members missing a sessionid)
+  - `hire` → `plan --member <alias>` (add a member to a teamfile)
+  - `clone` → `plan --from <teamfile>` (start a new teamfile from an existing one)
+  - `health` → `status <teamfile> [alias]` (team/member health view; the
+    `--json` output is versioned as `STATUS_JSON_SCHEMA_VERSION=1`)
+  - `team`, `session`, `session-index`, `restore`, `restore-teamfile`,
+    `diagnose`, `diagnose-prompt` → removed with no replacement (their duties
+    are covered by direct teamfile edits + `lint`, or were diagnostics
+    subsumed by `status`)
+- **New interactive `plan`.** Run with no flags, `plan` walks a wizard that
+  writes a teamfile; with flags it is fully non-interactive. The golden path is
+  now `lmctl plan` → edit the teamfile → `lmctl lint` → `lmctl seed`.
+- **New `status` modes.** `status <teamfile>` shows every member of a teamfile;
+  `status <teamfile> <alias>` drills into one member, absorbing the old
+  `health` verb.
+- **`pi` is now a first-class provider binary** (`PROVIDER_BINARIES`).
+- **The `@lmctl/next-agent` `diagnose-prompt` bin shim is removed.** The
+  passthrough shim no longer exposes that verb; `@lmctl/next-agent` consumers
+  get the same consolidated surface.
+
+## Unreleased
+
+- **The published package no longer bundles the `serve` daemon, job queue, or
+  workflow engine.** `@lmctl-ai/lmctl` was designed to ship prompt
+  self-orchestration only — no daemon, no queue — but every version published
+  to date (0.1.0 through 0.1.341) actually contained the full daemon/queue/
+  engine code in `dist/cli/index.js`; the `serve` command's `--no-daemon` flag
+  was a runtime opt-out, not a build-time exclusion. This is a packaging-scope
+  defect, not a reachability one: the daemon binds to loopback by default and
+  requires an explicit auth token for any non-loopback bind or for its HTTP
+  API even on loopback, so nothing shipped that a normal install exposed on a
+  network by default. Starting with this release, the public bundle has the
+  `serve` dispatch's target module graph removed at build time; the CLI's own
+  `serve` command is gone from this package (an internal build used for our
+  own daemon deployment keeps it). No prior version has been unpublished or
+  deprecated — this entry exists so the history says what changed rather than
+  quietly becoming correct.
 - **A live `lmctl chat` dispatch no longer orphans the receiver's own background
   work — and its duration is now subtree-bound, not turn-bound.** Previously the
   provider transport closed the agent's stdin the instant its turn produced

package/README.md

@@ -44,9 +44,7 @@ lmctl status
 lmctl plan ./my-project
 lmctl lint ./my-project/my-project.lmctl
 lmctl seed ./my-project/my-project.lmctl
-lmctl chat ./my-project/my-project.lmctl Lead "Inspect the repo and propose next steps."
-lmctl mail sent --json
-lmctl mail sent --limit 100 --cursor "$NEXT_CURSOR" --json
+lmctl prompt ./my-project/my-project.lmctl Lead "Inspect the repo and propose next steps."
 ```
 
 `lmctl status` is team/SELF-scoped: from a member session it shows your
@@ -55,23 +53,21 @@ From a plain operator shell it shows a DB-wide summary.
 Use `lmctl status --json` for the full machine-readable view; `--project` and
 `--web` are not status flags.
 
-Start with the plan -> lint -> seed -> chat flow. lmctl does not manage
+Start with the plan -> edit -> lint -> seed -> prompt flow. `lmctl plan` runs an
+interactive wizard when invoked without flags, or writes the teamfile directly
+from flags; edit the generated teamfile, then `lint` and `seed` it. lmctl does not manage
 foreground vs. background execution for provider turns or shell commands; use
 the provider or host runtime's own background/task surface when you need async
 execution.
-Daemon, web, and API surfaces are secondary integration paths; the CLI is the
-golden path for onboarding.
-To inspect the daemon, run `lmctl api daemon state` while `lmctl serve` is
-running; authenticate with `LMCTL_API_TOKEN=<token>` or start local serve with
-`--insecure-no-auth`.
+The CLI is the golden path for onboarding; lmctl is stateless (the former
+daemon/web/API surface moved to the separate lmbee product).
 Lintable teamfile templates for research, writing, and review teams ship in the
 [examples gallery](https://lmctl.com/examples/).
 
-`lmctl` top-level subcommands are: `api`, `chat`, `serve`, `device`,
-`team`, `terminal`, `tail`, `health`, `cost`, `session`, `session-index`, `restore`, `restore-teamfile`,
-`ls`, `lint`, `seed`, `hire`,
-`refresh`, `clone`, `plan`, `diagnose`, `diagnose-prompt`,
-`mail`, `status`, `running`, and `db`. Run `lmctl <command> --help` for details on each, and see the
+`lmctl` top-level subcommands are: `batch`, `chat`, `cost`, `db`,
+`export`, `lint`, `ls`, `plan`, `prompt`, `ratelimit`, `running`,
+`script`, `seed`, `status`, `tail`, and `terminal`. Run `lmctl <command> --help` for details on
+each, and see the
 [documentation](https://lmctl.com/lmctl) to define your first team.
 
 Abstract managed-team providers retain the config keys
@@ -85,43 +81,7 @@ sqlite3 ~/.lmctl/state.db "INSERT OR REPLACE INTO lmctl_config(key,value,updated
 sqlite3 ~/.lmctl/state.db "INSERT OR REPLACE INTO lmctl_config(key,value,updated_at) VALUES('workspace.provider3','gemini',strftime('%Y-%m-%dT%H:%M:%fZ','now'));"
 ```
 
-`lmctl mail` exposes low-level, read-only/record-only primitives over the
-event-log substrate: list pending or sent messages, read event history/
-content, inspect historical provider evidence and causal lineage, and declare
-handling/acknowledgement. `mail pending` defaults to messages addressed to
-SELF when run from a member session; an explicit `--receiver` overrides that
-scope. From a plain operator shell with no SELF marker, it retains the
-fleet-wide discovery view.
-Missing causal capture is shown as an explicit tree root rather than guessed
-from timestamps.
-For every `mail` subcommand, `--json` is the stable external contract and
-includes `schema_version: 1`; default text output is for humans and may change.
-JSON failures use `{status:"error", errorKind, message}`. The unbounded `sent`
-and `tree` scans are cursor-paginated, with 100 results by default and 1,000
-maximum; pass `next_cursor` back through `--cursor` until it is `null`.
-`schema_version` is the authoritative output-contract version; the separate
-`schema: "event-log-v1"` field identifies the backing substrate.
-
-Stable top-level success fields are:
-
-- `sent`: `schema_version`, `schema`, `identity`, `filters`, `messages`,
-  `next_cursor`.
-- `history`: `schema_version`, `schema`, `message_id`, `events`.
-- `seen`: `schema_version`, `schema`, `message_id`, `receiver`, `db`,
-  `transcript`, `answered`.
-- `pending`: `schema_version`, `schema`, `filters`, `messages`.
-- `read`: `schema_version`, `schema`, `message_id`, `message_type`,
-  `in_reply_to`, `sender`, `receiver`, `content`, `pending_state`.
-- `deliver`: `schema_version`, `schema`, `message_id`, `message_type`,
-  `receiver`, `outcome`, `provider_status`, `message`.
-- `handle`: `schema_version`, `schema`, `message_id`, `handling_by`,
-  `handling_at`, `appended`.
-- `ack`: `schema_version`, `schema`, `message_id`, `acked_by`, `appended`,
-  `pending_outcome`.
-- `tree`: `schema_version`, `schema`, `root`, `since`, `since_timestamp`,
-  `sent`, `next_cursor`, `orphan_policy`, `incomplete_capture_note`.
-
-- `lmctl chat <teamfile> <alias> "message"` drives or messages a member. If a member-origin target is busy, chat queues the message for that sender->receiver lane; the next chat **from that same sender** to that receiver, when it is free, delivers the queued lane plus the new message. The lane key is `(sender, receiver)`: **a chat to that receiver from anyone else does not flush your queued mail.** If you go idle waiting for the reply you queued, nothing else will unblock it — you are the only party whose next chat flushes that lane.
+- `lmctl prompt <teamfile> <alias> "prompt"` sends a prompt to a member and prints its answer. If a member-origin target is busy, the prompt is queued for that sender->receiver lane; the next prompt **from that same sender** to that receiver, when it is free, delivers the queued lane plus the new prompt. The lane key is `(sender, receiver)`: **a prompt to that receiver from anyone else does not flush your queued prompts.** If you go idle waiting for the answer you queued, nothing else will unblock it — you are the only party whose next prompt flushes that lane.
 
 ## Model selection
 

package/data/price-aliases.json

@@ -0,0 +1 @@
+{"schema_version":1,"notes":["First-party aliases cover harness strings that do not resolve mechanically to a catalog identity.","Consumers must exact-match aliases case-insensitively; table entries win over canonical_slug.","new_model means new to this catalog, not necessarily newly released by a vendor.","This table is where manual exception maintenance lives; adding support for a new special case means adding one entry.","There is no precedence hierarchy by design: this single table is the canonical exception home."],"standards":[{"standard_slug":"moonshot/kimi-k2.8-preview","display_name":"Kimi K2.8 Preview","vendor":"moonshot","catalog_record_id":"manual:kimi-code/kimi-for-coding","note":"Estimated subscription scenario; no first-party price."},{"standard_slug":"moonshot/kimi-k2.5","display_name":"Kimi K2.5","vendor":"moonshot","catalog_record_id":"litellm:moonshot/kimi-k2.5","note":"Direct Moonshot reference row; qwen/DashScope route is a separate billing path."},{"standard_slug":"moonshot/kimi-k2.7-code-highspeed","display_name":"Kimi K2.7 Code HighSpeed","vendor":"moonshot","catalog_record_id":"manual:kimi-code/kimi-for-coding-highspeed","note":"Estimated Code scenario; Code and platform billing are separate."},{"standard_slug":"moonshot/kimi-k3","display_name":"Kimi K3","vendor":"moonshot","catalog_record_id":"manual:kimi-code/k3","note":"Operator-asserted price-proxy identity; the historical binding refusal remains documented in each alias."},{"standard_slug":"deepseek/deepseek-flash","display_name":"DeepSeek Flash","vendor":"deepseek","catalog_record_id":"litellm:deepseek/deepseek-flash","note":null},{"standard_slug":"deepseek/deepseek-v4-pro","display_name":"DeepSeek V4 Pro","vendor":"deepseek","catalog_record_id":"litellm:deepseek/deepseek-v4-pro","note":null},{"standard_slug":"google/gemini-3.8-flash","display_name":"Gemini 3.8 Flash","vendor":"google","catalog_record_id":"litellm:gemini-3.8-flash","note":null},{"standard_slug":"google/gemini-3.7-flash","display_name":"Gemini 3.7 Flash","vendor":"google","catalog_record_id":"litellm:gemini-3.7-flash","note":null},{"standard_slug":"google/gemini-3.6-flash","display_name":"Gemini 3.6 Flash","vendor":"google","catalog_record_id":"litellm:gemini-3.6-flash","note":null},{"standard_slug":"google/gemini-3.1-pro-preview","display_name":"Gemini 3.1 Pro Preview","vendor":"google","catalog_record_id":"litellm:gemini/gemini-3.1-pro-preview","note":"Harness mapping is asserted pending native-owner evidence."},{"standard_slug":"anthropic/claude-opus-4-6","display_name":"Claude Opus 4.6","vendor":"anthropic","catalog_record_id":"litellm:claude-opus-4-6","note":null},{"standard_slug":"anthropic/claude-sonnet-4-6","display_name":"Claude Sonnet 4.6","vendor":"anthropic","catalog_record_id":"litellm:claude-sonnet-4-6","note":null},{"standard_slug":"dashscope/qwen3.5-plus","display_name":"Qwen 3.5 Plus","vendor":"dashscope","catalog_record_id":"litellm:dashscope/qwen3.5-plus","note":"Identity established; catalog pricing may be null."},{"standard_slug":"dashscope/qwen3.6-plus","display_name":"Qwen 3.6 Plus","vendor":"dashscope","catalog_record_id":"manual:dashscope/qwen3.6-plus","note":"Vendor-sourced manual row; official pricing is tiered and absent from the upstream row."},{"standard_slug":"z-ai/glm-5.2","display_name":"GLM-5.2","vendor":"z-ai","catalog_record_id":"litellm:zai/glm-5.2","note":"Plan-route identity; catalog rate is counterfactual standard pricing."},{"standard_slug":"z-ai/glm-5.1","display_name":"GLM-5.1","vendor":"z-ai","catalog_record_id":"litellm:zai/glm-5.1","note":"Plan-route identity; catalog rate is counterfactual standard pricing."},{"standard_slug":"z-ai/glm-4.7","display_name":"GLM-4.7","vendor":"z-ai","catalog_record_id":"litellm:zai/glm-4.7","note":"Plan-route identity; catalog rate is counterfactual standard pricing."},{"standard_slug":"z-ai/glm-5","display_name":"GLM-5","vendor":"z-ai","catalog_record_id":"litellm:zai/glm-5","note":"DashScope route identity; direct zai row is the counterfactual reference."},{"standard_slug":"z-ai/glm-5-turbo","display_name":"GLM-5 Turbo","vendor":"z-ai","catalog_record_id":"openrouter:z-ai/glm-5-turbo","note":"Z-AI coding plan identity; OpenRouter rate is a counterfactual reference because no direct zai row exists."},{"standard_slug":"z-ai/glm-4.5-air","display_name":"GLM-4.5 Air","vendor":"z-ai","catalog_record_id":"litellm:zai/glm-4.5-air","note":"Z-AI coding plan identity; standard API rate is counterfactual."},{"standard_slug":"openrouter/qwen/qwen3.6-plus","display_name":"Qwen 3.6 Plus (OpenRouter)","vendor":"openrouter","catalog_record_id":"openrouter:qwen/qwen3.6-plus","note":"OpenRouter gateway-priced route; actual opencode sub-provider billing path."},{"standard_slug":"openai/gpt-5.6-sol","display_name":"GPT-5.6 Sol","vendor":"openai","catalog_record_id":"litellm:gpt-5.6-sol","note":"Direct OpenAI billing route; plan route pricing is counterfactual."},{"standard_slug":"openai/gpt-5.6-luna","display_name":"GPT-5.6 Luna","vendor":"openai","catalog_record_id":"litellm:gpt-5.6-luna","note":"Direct OpenAI billing route."},{"standard_slug":"openai/gpt-6-astra","display_name":"GPT-6 Astra","vendor":"openai","catalog_record_id":"litellm:gpt-6-astra","note":"Direct OpenAI billing route."},{"standard_slug":"openai/gpt-5.3-codex-spark","display_name":"GPT-5.3 Codex Spark","vendor":"openai","catalog_record_id":"litellm:chatgpt/gpt-5.3-codex-spark","note":"Established Codex subscription SKU; OpenAI marks pricing as research preview/not final, so the catalog row remains null-priced."},{"standard_slug":"anthropic/claude-haiku-4.5","display_name":"Claude Haiku 4.5","vendor":"anthropic","catalog_record_id":"openrouter:anthropic/claude-haiku-4.5","note":"Vendor identity; route-specific plan price is unavailable, so OpenRouter is a reference row."},{"standard_slug":"anthropic/claude-opus-4.5","display_name":"Claude Opus 4.5","vendor":"anthropic","catalog_record_id":"openrouter:anthropic/claude-opus-4.5","note":"Vendor identity; route-specific plan price is unavailable, so OpenRouter is a reference row."},{"standard_slug":"anthropic/claude-opus-4.7","display_name":"Claude Opus 4.7","vendor":"anthropic","catalog_record_id":"openrouter:anthropic/claude-opus-4.7","note":"Vendor identity; route-specific plan price is unavailable, so OpenRouter is a reference row."},{"standard_slug":"anthropic/claude-opus-4.8","display_name":"Claude Opus 4.8","vendor":"anthropic","catalog_record_id":"openrouter:anthropic/claude-opus-4.8","note":"Vendor identity; route-specific plan price is unavailable, so OpenRouter is a reference row."},{"standard_slug":"anthropic/claude-sonnet-4.5","display_name":"Claude Sonnet 4.5","vendor":"anthropic","catalog_record_id":"openrouter:anthropic/claude-sonnet-4.5","note":"Vendor identity; route-specific plan price is unavailable, so OpenRouter is a reference row."},{"standard_slug":"anthropic/claude-sonnet-5","display_name":"Claude Sonnet 5","vendor":"anthropic","catalog_record_id":"litellm:claude-sonnet-5","note":"Direct Anthropic reference row; Copilot plan rate is counterfactual."},{"standard_slug":"deepseek/deepseek-chat","display_name":"DeepSeek Chat","vendor":"deepseek","catalog_record_id":"litellm:deepseek/deepseek-chat","note":"Direct DeepSeek billing route."},{"standard_slug":"deepseek/deepseek-reasoner","display_name":"DeepSeek Reasoner","vendor":"deepseek","catalog_record_id":"litellm:deepseek/deepseek-reasoner","note":"Direct DeepSeek billing route."},{"standard_slug":"deepseek/deepseek-v4-flash","display_name":"DeepSeek V4 Flash","vendor":"deepseek","catalog_record_id":"litellm:deepseek/deepseek-v4-flash","note":"Direct DeepSeek billing route."},{"standard_slug":"google/gemini-2.5-pro","display_name":"Gemini 2.5 Pro","vendor":"google","catalog_record_id":"litellm:gemini-2.5-pro","note":"Direct Google billing route; Copilot plan rate is counterfactual."},{"standard_slug":"google/gemini-3-flash-preview","display_name":"Gemini 3 Flash Preview","vendor":"google","catalog_record_id":"litellm:gemini-3-flash-preview","note":"Direct Google billing route; Copilot plan rate is counterfactual."},{"standard_slug":"google/gemini-3.5-flash","display_name":"Gemini 3.5 Flash","vendor":"google","catalog_record_id":"litellm:gemini-3.5-flash","note":"Direct Google billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-4.1","display_name":"GPT-4.1","vendor":"openai","catalog_record_id":"litellm:gpt-4.1","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-4.1-2025-04-14","display_name":"GPT-4.1 (2025-04-14)","vendor":"openai","catalog_record_id":"litellm:gpt-4.1-2025-04-14","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-4o-mini","display_name":"GPT-4o Mini","vendor":"openai","catalog_record_id":"litellm:gpt-4o-mini","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-4o-mini-2024-07-18","display_name":"GPT-4o Mini (2024-07-18)","vendor":"openai","catalog_record_id":"litellm:gpt-4o-mini-2024-07-18","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5-mini","display_name":"GPT-5 Mini","vendor":"openai","catalog_record_id":"litellm:gpt-5-mini","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.2","display_name":"GPT-5.2","vendor":"openai","catalog_record_id":"litellm:gpt-5.2","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.2-codex","display_name":"GPT-5.2 Codex","vendor":"openai","catalog_record_id":"litellm:gpt-5.2-codex","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.3-codex","display_name":"GPT-5.3 Codex","vendor":"openai","catalog_record_id":"litellm:gpt-5.3-codex","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.4","display_name":"GPT-5.4","vendor":"openai","catalog_record_id":"litellm:gpt-5.4","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.4-mini","display_name":"GPT-5.4 Mini","vendor":"openai","catalog_record_id":"litellm:gpt-5.4-mini","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.4-nano","display_name":"GPT-5.4 Nano","vendor":"openai","catalog_record_id":"litellm:gpt-5.4-nano","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"openai/gpt-5.5","display_name":"GPT-5.5","vendor":"openai","catalog_record_id":"litellm:gpt-5.5","note":"Direct OpenAI billing route; Copilot plan rate is counterfactual."},{"standard_slug":"qwen/qwen3-coder","display_name":"Qwen3 Coder","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3-coder","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3-coder-flash","display_name":"Qwen3 Coder Flash","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3-coder-flash","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3-coder-plus","display_name":"Qwen3 Coder Plus","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3-coder-plus","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.5-397b-a17b","display_name":"Qwen3.5 397B A17B","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.5-397b-a17b","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.5-9b","display_name":"Qwen3.5 9B","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.5-9b","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.5-plus-20260420","display_name":"Qwen3.5 Plus 2026-04-20","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.5-plus-20260420","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.6-flash","display_name":"Qwen3.6 Flash","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.6-flash","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.6-max-preview","display_name":"Qwen3.6 Max Preview","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.6-max-preview","note":"OpenRouter gateway-priced actual route."},{"standard_slug":"qwen/qwen3.7-plus","display_name":"Qwen3.7 Plus","vendor":"qwen","catalog_record_id":"openrouter:qwen/qwen3.7-plus","note":"OpenRouter gateway-priced actual route."}],"aliases":[{"alias":"kimi-code/kimi-for-coding","standard_slug":"moonshot/kimi-k2.8-preview","evidence":"established","source_url":"https://www.kimi.com/code/docs/en/kimi-code/models.html","note":"Kimi Code docs name K2.8 Preview (up to 1M context); subscription-billed and no first-party price.","added_at":"2026-09-23"},{"alias":"kimi-code/kimi-for-coding-highspeed","standard_slug":"moonshot/kimi-k2.7-code-highspeed","evidence":"established","source_url":"https://www.kimi.com/code/docs/en/kimi-code/models.html","note":"Docs name K2.7 Code HighSpeed; Code and platform billing remain separate, so this is a scenario identity.","added_at":"2026-09-23"},{"alias":"kimi-code/k3","standard_slug":"moonshot/kimi-k3","evidence":"asserted","source_url":null,"source":"operator 2026-09-25","note":"Asserted for price-proxy purposes only. Historical refusal remains: two documented sweeps (2026-09-13 and 2026-09-21), product-independence statement, and ruling 29144f0 left Code↔Platform binding unestablished; the identity question stays closed for this estimate.","added_at":"2026-09-25"},{"alias":"kimi-code/k3-256k","standard_slug":"moonshot/kimi-k3","evidence":"asserted","source_url":null,"source":"operator 2026-09-25","note":"Asserted for price-proxy purposes only; k3-256k is the K3 256K variant per docs. Historical refusal remains: two documented sweeps, product-independence statement, and ruling 29144f0 left Code↔Platform binding unestablished; the identity question stays closed for this estimate.","added_at":"2026-09-25"},{"alias":"dsh/deepseek-flash","standard_slug":"deepseek/deepseek-flash","evidence":"established","source_url":"https://api-docs.deepseek.com/updates/","note":"Official 2026-09-10 changelog names deepseek-flash as the current V4.1 Flash model.","added_at":"2026-09-23"},{"alias":"dsh/deepseek-v4-flash","standard_slug":"deepseek/deepseek-flash","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"Official pricing footnote says the retired alias routes to V4.1-Flash and is billed at Flash price.","added_at":"2026-09-23"},{"alias":"dsh/deepseek-v4-pro","standard_slug":"deepseek/deepseek-v4-pro","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"Current official DeepSeek model ID.","added_at":"2026-09-23"},{"alias":"agy/gemini-3.8-flash","standard_slug":"google/gemini-3.8-flash","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/models","note":"Exact official model ID.","added_at":"2026-09-23"},{"alias":"agy/gemini-3.7-flash","standard_slug":"google/gemini-3.7-flash","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/models","note":"Exact official model ID.","added_at":"2026-09-23"},{"alias":"agy/gemini-3.6-flash","standard_slug":"google/gemini-3.6-flash","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/models","note":"Exact official model ID.","added_at":"2026-09-23"},{"alias":"agy/gemini-pro-default","standard_slug":"google/gemini-3.1-pro-preview","evidence":"asserted","source_url":null,"source":"lmprice #241","note":"Catalog and pricing-rule half is established, but the Antigravity harness half is unproven; pending native-owner evidence.","added_at":"2026-09-23"},{"alias":"agy/claude-opus-4-6-thinking","standard_slug":"anthropic/claude-opus-4-6","evidence":"established","source_url":"https://platform.claude.com/docs/en/build-with-claude/thinking-steering-and-cost","note":"Thinking steering and cost guidance documents thinking as output; Antigravity lists Claude Opus 4.6 (thinking).","added_at":"2026-09-23"},{"alias":"agy/claude-sonnet-4-6","standard_slug":"anthropic/claude-sonnet-4-6","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"Exact model identity.","added_at":"2026-09-23"},{"alias":"agy/gemini-3.8-flash-tiered","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"Harness-specific tier label; no official SKU or identity evidence.","added_at":"2026-09-23"},{"alias":"agy/gemini-3.1-pro-low","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"Harness effort label; no official SKU or identity evidence.","added_at":"2026-09-23"},{"alias":"qwen/coder-model","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"Harness alias; no vendor identity evidence.","added_at":"2026-09-23"},{"alias":"opencode/glm-5.1","standard_slug":"z-ai/glm-5.1","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Exact GLM model name on a plan route; pricing is counterfactual.","added_at":"2026-09-23"},{"alias":"opencode/glm-4.7","standard_slug":"z-ai/glm-4.7","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Exact GLM model name on a plan route; pricing is counterfactual.","added_at":"2026-09-23"},{"alias":"zai-coding-plan/GLM-5.2","standard_slug":"z-ai/glm-5.2","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"z-ai-coding/GLM-5.2","standard_slug":"z-ai/glm-5.2","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"zai/GLM-5.2","standard_slug":"z-ai/glm-5.2","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"zai-coding-plan/GLM-5.1","standard_slug":"z-ai/glm-5.1","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"z-ai-coding/GLM-5.1","standard_slug":"z-ai/glm-5.1","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"zai/GLM-5.1","standard_slug":"z-ai/glm-5.1","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"zai-coding-plan/GLM-4.7","standard_slug":"z-ai/glm-4.7","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"z-ai-coding/GLM-4.7","standard_slug":"z-ai/glm-4.7","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"zai/GLM-4.7","standard_slug":"z-ai/glm-4.7","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Plan route; standard API rate is counterfactual.","added_at":"2026-09-23"},{"alias":"agy/gemini 3.8 flash (high)","standard_slug":"google/gemini-3.8-flash","evidence":"established","source_url":"https://antigravity.google/docs/models","note":"Antigravity model enumeration names Gemini 3.8 Flash (High); effort label is not a priced SKU, following the gemini-3.1-pro-low precedent.","added_at":"2026-09-24"},{"alias":"agy/gemini 3.8 flash (medium)","standard_slug":"google/gemini-3.8-flash","evidence":"established","source_url":"https://antigravity.google/docs/models","note":"Antigravity model enumeration names Gemini 3.8 Flash (Medium); effort label is not a priced SKU, following the gemini-3.1-pro-low precedent.","added_at":"2026-09-24"},{"alias":"agy/gemini 3.1 pro (high)","standard_slug":"google/gemini-3.1-pro-preview","evidence":"established","source_url":"https://antigravity.google/docs/models","note":"Antigravity enumerates Gemini 3.1 Pro High; High is an effort label, not a priced SKU, and Google publishes Gemini 3.1 Pro Preview.","added_at":"2026-09-24"},{"alias":"agy/claude opus 4.6 (thinking)","standard_slug":"anthropic/claude-opus-4-6","evidence":"established","source_url":"https://antigravity.google/docs/models","note":"Antigravity enumerates Claude Opus 4.6 (thinking); Anthropic documents thinking tokens as output, so thinking is not a separate priced SKU.","added_at":"2026-09-24"},{"alias":"agy/gemini-3.1-pro","standard_slug":"google/gemini-3.1-pro-preview","evidence":"asserted","source_url":"https://antigravity.google/docs/models","source":"Antigravity /docs/models enumeration","note":"Bare harness slug asserted to the published Gemini 3.1 Pro Preview identity; weaker than the labeled enumeration form.","added_at":"2026-09-24"},{"alias":"qwen/glm-5","standard_slug":"z-ai/glm-5","evidence":"established","source_url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"DashScope hosts GLM-5; pin to the direct zai catalog row as the standard identity. The direct zai rate is a counterfactual reference for the qwen/DashScope route.","added_at":"2026-09-24"},{"alias":"qwen/kimi-k2.5","standard_slug":"moonshot/kimi-k2.5","evidence":"established","source_url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"DashScope hosts Kimi K2.5; pin to the direct moonshot catalog row as the standard identity. The direct moonshot rate is a counterfactual reference for the qwen/DashScope route.","added_at":"2026-09-24"},{"alias":"qwen/qwen3.5-plus","standard_slug":"dashscope/qwen3.5-plus","evidence":"established","source_url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"Native qwen routes to DashScope; official tiered pricing now fills the null catalog row.","added_at":"2026-09-24"},{"alias":"qwen/qwen3.6-plus","standard_slug":"dashscope/qwen3.6-plus","evidence":"established","source_url":"https://www.alibabacloud.com/help/en/model-studio/model-pricing","note":"Official DashScope model and tiered pricing; native qwen billing route.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.6-plus","standard_slug":"openrouter/qwen/qwen3.6-plus","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.6-plus","note":"OpenRouter sub-provider is the actual opencode billing path; gateway-priced actual, with route-rate oscillation caveat.","added_at":"2026-09-24"},{"alias":"codex/gpt-5.3-codex-spark","standard_slug":"openai/gpt-5.3-codex-spark","evidence":"established","source_url":"https://help.openai.com/en/articles/20001415-chatgpt-rate-card-enterprise-token-based-pricing","note":"Exact catalog row and established Codex subscription SKU; OpenAI marks all rates research preview/not final, so this is known-unpriced terminal.","added_at":"2026-09-24"},{"alias":"opencode/claude-haiku-4.5","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for this exact model, so no standard-price proxy is asserted.","added_at":"2026-09-24"},{"alias":"opencode/claude-opus-4.5","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for this exact model, so no standard-price proxy is asserted.","added_at":"2026-09-24"},{"alias":"opencode/claude-opus-4.6","standard_slug":"anthropic/claude-opus-4-6","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"GitHub Copilot plan billing; direct vendor rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/claude-opus-4.7","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for this exact model, so no standard-price proxy is asserted.","added_at":"2026-09-24"},{"alias":"opencode/claude-opus-4.8","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; only gateway/reference rows exist, not a direct vendor LiteLLM row.","added_at":"2026-09-24"},{"alias":"opencode/claude-sonnet-4.5","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for this exact model.","added_at":"2026-09-24"},{"alias":"opencode/claude-sonnet-4.6","standard_slug":"anthropic/claude-sonnet-4-6","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"GitHub Copilot plan billing; direct vendor rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/claude-sonnet-5","standard_slug":"anthropic/claude-sonnet-5","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"GitHub Copilot plan billing; direct vendor rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/deepseek-chat","standard_slug":"deepseek/deepseek-chat","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek sub-provider; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/deepseek-reasoner","standard_slug":"deepseek/deepseek-reasoner","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek sub-provider; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/deepseek-v4-flash","standard_slug":"deepseek/deepseek-v4-flash","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek sub-provider; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/deepseek-v4-pro","standard_slug":"deepseek/deepseek-v4-pro","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek sub-provider; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/deepseek/deepseek-flash","standard_slug":"deepseek/deepseek-flash","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek route-qualified form; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/deepseek/deepseek-v4-flash","standard_slug":"deepseek/deepseek-v4-flash","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek route-qualified form; direct vendor billing route.","added_at":"2026-09-24"},{"alias":"opencode/gemini-2.5-pro","standard_slug":"google/gemini-2.5-pro","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/pricing","note":"GitHub Copilot plan billing; direct Google rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gemini-3-flash-preview","standard_slug":"google/gemini-3-flash-preview","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/pricing","note":"GitHub Copilot plan billing; direct Google rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gemini-3.1-pro-preview","standard_slug":"google/gemini-3.1-pro-preview","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/pricing","note":"GitHub Copilot plan billing; direct Google rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gemini-3.5-flash","standard_slug":"google/gemini-3.5-flash","evidence":"established","source_url":"https://ai.google.dev/gemini-api/docs/pricing","note":"GitHub Copilot plan billing; direct Google rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/github-copilot/claude-opus-4.8","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for Claude Opus 4.8.","added_at":"2026-09-24"},{"alias":"opencode/github-copilot/claude-sonnet-5","standard_slug":"anthropic/claude-sonnet-5","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"GitHub Copilot plan billing; direct vendor rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/github-copilot/gpt-5.5","standard_slug":"openai/gpt-5.5","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/github-copilot/gpt-5.6-sol","standard_slug":"openai/gpt-5.6-sol","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/glm-4.5-air","standard_slug":"z-ai/glm-4.5-air","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Z-AI coding-plan route; direct zai rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/glm-5","standard_slug":"z-ai/glm-5","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Z-AI coding-plan route; direct zai rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/glm-5-turbo","standard_slug":"z-ai/glm-5-turbo","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Z-AI coding-plan route; OpenRouter rate is the counterfactual reference because no direct zai row exists.","added_at":"2026-09-24"},{"alias":"opencode/glm-5.2","standard_slug":"z-ai/glm-5.2","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Z-AI coding-plan route; direct zai rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-4.1","standard_slug":"openai/gpt-4.1","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-4.1-2025-04-14","standard_slug":"openai/gpt-4.1-2025-04-14","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-4o-mini","standard_slug":"openai/gpt-4o-mini","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-4o-mini-2024-07-18","standard_slug":"openai/gpt-4o-mini-2024-07-18","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5-mini","standard_slug":"openai/gpt-5-mini","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.2","standard_slug":"openai/gpt-5.2","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.2-codex","standard_slug":"openai/gpt-5.2-codex","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.3-codex","standard_slug":"openai/gpt-5.3-codex","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.4","standard_slug":"openai/gpt-5.4","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.4-mini","standard_slug":"openai/gpt-5.4-mini","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.4-nano","standard_slug":"openai/gpt-5.4-nano","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-5.5","standard_slug":"openai/gpt-5.5","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"opencode/gpt-6-astra","standard_slug":"openai/gpt-6-astra","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"OpenAI sub-provider; direct OpenAI billing route.","added_at":"2026-09-24"},{"alias":"opencode/mai-code-1-flash-picker","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no catalog row or vendor identity evidence exists for this picker id.","added_at":"2026-09-24"},{"alias":"opencode/openai/gpt-6-astra","standard_slug":"openai/gpt-6-astra","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"OpenAI route-qualified form; direct OpenAI billing route.","added_at":"2026-09-24"},{"alias":"opencode/openrouter/qwen/qwen3.7-plus","standard_slug":"qwen/qwen3.7-plus","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.7-plus","note":"OpenRouter route-qualified form; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3-coder","standard_slug":"qwen/qwen3-coder","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3-coder","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3-coder-flash","standard_slug":"qwen/qwen3-coder-flash","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3-coder-flash","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3-coder-plus","standard_slug":"qwen/qwen3-coder-plus","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3-coder-plus","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.5-397b-a17b","standard_slug":"qwen/qwen3.5-397b-a17b","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.5-397b-a17b","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.5-9b","standard_slug":"qwen/qwen3.5-9b","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.5-9b","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.5-plus-20260420","standard_slug":"qwen/qwen3.5-plus-20260420","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.5-plus-20260420","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.6-flash","standard_slug":"qwen/qwen3.6-flash","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.6-flash","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.6-max-preview","standard_slug":"qwen/qwen3.6-max-preview","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.6-max-preview","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/qwen/qwen3.7-plus","standard_slug":"qwen/qwen3.7-plus","evidence":"established","source_url":"https://openrouter.ai/qwen/qwen3.7-plus","note":"OpenRouter sub-provider; gateway-priced actual billing path.","added_at":"2026-09-24"},{"alias":"opencode/z-ai-coding/GLM-5.2","standard_slug":"z-ai/glm-5.2","evidence":"established","source_url":"https://docs.z.ai/guides/overview","note":"Z-AI coding-plan route; direct zai rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"lmplayer/deepseek/deepseek-v4-pro","standard_slug":"deepseek/deepseek-v4-pro","evidence":"established","source_url":"https://api-docs.deepseek.com/quick_start/pricing","note":"DeepSeek route; direct vendor billing path.","added_at":"2026-09-24"},{"alias":"lmplayer/github-copilot/claude-opus-4.8","standard_slug":null,"evidence":"unresolved","source_url":null,"note":"GitHub Copilot plan route; no direct vendor LiteLLM row exists for Claude Opus 4.8.","added_at":"2026-09-24"},{"alias":"lmplayer/github-copilot/claude-sonnet-5","standard_slug":"anthropic/claude-sonnet-5","evidence":"established","source_url":"https://platform.claude.com/docs/en/about-claude/models","note":"GitHub Copilot plan billing; direct vendor rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"lmplayer/github-copilot/gpt-5.5","standard_slug":"openai/gpt-5.5","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"GitHub Copilot plan billing; direct OpenAI rate is a standard-rate counterfactual.","added_at":"2026-09-24"},{"alias":"lmplayer/openai/gpt-6-astra","standard_slug":"openai/gpt-6-astra","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"OpenAI route; direct vendor billing path.","added_at":"2026-09-24"},{"alias":"pi/openai/gpt-5.6-luna","standard_slug":"openai/gpt-5.6-luna","evidence":"established","source_url":"https://platform.openai.com/docs/pricing","note":"OpenAI route; direct vendor billing path.","added_at":"2026-09-24"}]}