diff --git a/CHANGELOG.md b/CHANGELOG.md index 6ffbf98..60479ad 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -2,6 +2,15 @@ ## Unreleased +- Refresh the static capability snapshot to `command-code@1.87.2`, which removes image, reasoning, and effort metadata for the retired free MiniMax M3.1 Flash and Ornith 1.5 122B variants. Paid variants and pricing are unchanged. + +- Refresh the static capability snapshot to `command-code@1.87.1`, which publishes no model metadata changes, and re-verify display pricing on 2026-10-22 for all 94 advertised models. Both free promotions ended on 2026-10-22 and the API dropped the models, so remove the `minimax/minimax-m3.1-flash:free` and `ornith/ornith-1.5-122b:free` display prices, their temporary-pricing guards, and their fixture entries; `TEMPORARY_PRICING` is empty until the pricing page publishes another dated promotion. Paid MiniMax M3.1 Flash ($0.15 input / $0.60 output / $0.015 cache read) and paid Ornith 1.5 122B ($0.16 input / $0.64 output / $0.02 cache read) rates are unchanged, and Beam's capacity-limited free pricing still has no published expiry. + +- Refresh the static capability snapshot to `command-code@1.87.0` and re-verify display pricing on 2026-10-21 for all 96 advertised models. MiniMax M3.1 Flash (paid and free) now supports image input, reasoning, and `low`/`medium`/`high`/`xhigh`/`max` efforts. Beam now supports reasoning with the same efforts and a 131,072-token output limit. Add paid and free Ornith 1.5 122B with image input and reasoning, without selectable efforts; paid rates are $0.16 input / $0.64 output / $0.02 cache read per million tokens, with no cache-write charge. Remove retired Cat Jam capability metadata. Extend the free MiniMax promotion to 2026-10-22T10:45:00Z and track the free Ornith promotion until 2026-10-22T06:00:00Z; both date-only guards require review from 2026-10-22 at 00:00 UTC, before either promotion ends. Existing paid rates and Beam's capacity-limited free pricing are unchanged. + +- Re-verify display pricing on 2026-10-20: add paid and free `minimax/minimax-m3.1-flash` entries and remove retired `stealth/cat-jam:free` pricing. The free MiniMax promotion ends on 2026-10-21 at 10:45 UTC; require a pricing recheck before that UTC day starts. The API now advertises 94 models. The CLI remains at `1.84.0` without MiniMax M3.1 Flash capability metadata, so both variants retain conservative text-only, non-reasoning defaults until upstream publishes it. + +- Refresh the static capability snapshot to `command-code@1.84.0`, which publishes no model metadata changes, and add reviewed display pricing for the new free `reflection/beam:free` model, verified against the official pricing page on 2026-10-17. Beam is free only while capacity lasts, with no published expiry, and it has no CLI capability metadata yet, so it keeps the conservative text-only and non-reasoning defaults with a 65,536-token output limit. Reasoning and effort selection will become available after upstream publishes the required CLI metadata and the catalog snapshot is refreshed. The reviewed snapshot now covers all 93 advertised models. - Refresh the static capability snapshot to `command-code@1.83.0`: add image input, reasoning, and selectable `low`/`medium`/`high` efforts for `meta/muse-spark-open` and `nvidia/nemotron-3.5-super-vl`; publish reasoning for the free `stealth/cat-jam:free` preview with `low`/`medium`/`high`/`xhigh`/`max` efforts and a 131,072-token output limit. Retire `inclusionai/ling-3.0-flash-sante:free` and drop its reasoning flag, output limit, and display-price entry now that the API no longer advertises it. No prices changed; the reviewed snapshot now covers all 92 advertised models. ## 0.7.8 - 2026-10-15 diff --git a/README.md b/README.md index 34a32da..aeaf458 100644 --- a/README.md +++ b/README.md @@ -151,7 +151,7 @@ The following environment variables are intended for tests, local mocks, and com ## Image input -The provider advertises image input only for models marked with the `image` input modality in the official Command Code CLI model catalog. The capability snapshot currently follows `command-code@1.83.0`; unknown models default to text-only until their upstream metadata is reviewed. A daily GitHub Actions job synchronizes the CLI version, image capabilities, reasoning flags, reasoning efforts, and model-specific output limits with the latest published CLI package, also dropping manual effort overrides that upstream has published itself, and opens or updates a reviewable pull request when they change. Pricing remains manually reviewed because temporary promotions and long-context tiers require explicit review. +The provider advertises image input only for models marked with the `image` input modality in the official Command Code CLI model catalog. The capability snapshot currently follows `command-code@1.87.2`; unknown models default to text-only until their upstream metadata is reviewed. A daily GitHub Actions job synchronizes the CLI version, image capabilities, reasoning flags, reasoning efforts, and model-specific output limits with the latest published CLI package, also dropping manual effort overrides that upstream has published itself, and opens or updates a reviewable pull request when they change. Pricing remains manually reviewed because temporary promotions and long-context tiers require explicit review. For vision-capable models, Pi's native provider adapters forward image blocks from user messages and tool results using the documented OpenAI or Anthropic message schema. Unknown and text-only models remain marked text-only in Pi. diff --git a/src/commandcode-catalog.ts b/src/commandcode-catalog.ts index 8c5e08f..1b7aa79 100644 --- a/src/commandcode-catalog.ts +++ b/src/commandcode-catalog.ts @@ -1,4 +1,4 @@ -export const COMMAND_CODE_CLI_VERSION = "1.83.0" +export const COMMAND_CODE_CLI_VERSION = "1.87.2" export type CommandCodeInputType = "text" | "image" export type CommandCodeReasoningEffort = @@ -11,7 +11,7 @@ export type CommandCodeReasoningEffort = | "max" /** - * Generated from command-code@1.83.0 by `npm run sync:commandcode-catalog`. + * Generated from command-code@1.87.2 by `npm run sync:commandcode-catalog`. * Do not edit manually. */ export const MODEL_INPUT_MODALITIES: Readonly> = { @@ -54,6 +54,7 @@ export const MODEL_INPUT_MODALITIES: Readonly> = { "meta/muse-spark-1.3-contributor": true, "meta/muse-spark-open": true, "meta/muse-wonder-1.0": true, + "minimax/minimax-m3.1-flash": true, "MiniMaxAI/MiniMax-M3": true, "mistral/mistral-large-4": true, "moonshotai/Kimi-K2.7-Code": true, @@ -138,6 +141,7 @@ export const MODEL_REASONING: Readonly> = { "moonshotai/Kimi-K3": true, "nvidia/nemotron-3-ultra-550b-a55b": true, "nvidia/nemotron-3.5-super-vl": true, + "ornith/ornith-1.5-122b": true, "poolside/laguna-s-2.1-free": true, "Qwen/Qwen3.6-Max-Preview": true, "Qwen/Qwen3.6-Plus": true, @@ -150,8 +154,8 @@ export const MODEL_REASONING: Readonly> = { "Qwen/Qwen3.8-Max-0902": true, "Qwen/Qwen3.8-Max-Prime": true, "Qwen/Qwen3.8-Omni-Flash": true, + "reflection/beam:free": true, "sakana/fugu-ultra": true, - "stealth/cat-jam:free": true, "stealth/glyph-cluster:free": true, "stepfun/Step-3.5-Flash": true, "stepfun/Step-3.7-Flash": true, @@ -212,6 +216,7 @@ export const MODEL_EFFORTS: Readonly> = { "Qwen/Qwen3.8-27B": 32_768, "Qwen/Qwen3.8-Max-Prime": 131_072, "Qwen/Qwen3.8-Omni-Flash": 131_072, - "stealth/cat-jam:free": 131_072, + "reflection/beam:free": 131_072, "stealth/glyph-cluster:free": 256_000, "z-ai/glm-5.3-flash": 131_072, "z-ai/glm-5.3-flashx": 131_072, diff --git a/src/pricing.ts b/src/pricing.ts index 24c9f0b..0cab7d0 100644 --- a/src/pricing.ts +++ b/src/pricing.ts @@ -20,7 +20,7 @@ export interface TemporaryPricing { } export const PRICING_SOURCE_URL = "https://commandcode.ai/docs/resources/pricing-limits" -export const PRICING_LAST_VERIFIED = "2026-10-15" +export const PRICING_LAST_VERIFIED = "2026-10-22" export const ZERO_MODEL_COST: CommandCodeModelCost = { input: 0, @@ -39,12 +39,13 @@ export const ZERO_MODEL_COST: CommandCodeModelCost = { */ export const MODEL_COSTS: Readonly> = { // Free models - // Free while the stealth preview lasts; no fixed expiry is published. - "stealth/cat-jam:free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, "poolside/laguna-s-2.1-free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, "stealth/glyph-cluster:free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, + // Free while capacity lasts; no fixed expiry is published. + "reflection/beam:free": { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }, // Open and open-weight models + "ornith/ornith-1.5-122b": { input: 0.16, output: 0.64, cacheRead: 0.02, cacheWrite: 0 }, "inclusionai/ling-3.1-flash": { input: 0.3, output: 0.9, cacheRead: 0.06, cacheWrite: 0 }, "mistral/mistral-large-4": { input: 1.36, output: 4.18, cacheRead: 0.14, cacheWrite: 0 }, "tencent/hy3-paid": { input: 0.14, output: 0.58, cacheRead: 0.035, cacheWrite: 0 }, @@ -66,6 +67,7 @@ export const MODEL_COSTS: Readonly> = { "zai-org/GLM-5.2-Fast": { input: 3, output: 10.25, cacheRead: 0.5, cacheWrite: 0 }, "zai-org/GLM-5.1": { input: 1.4, output: 4.4, cacheRead: 0.26, cacheWrite: 0 }, "zai-org/GLM-5": { input: 1, output: 3.2, cacheRead: 0.2, cacheWrite: 0 }, + "minimax/minimax-m3.1-flash": { input: 0.15, output: 0.6, cacheRead: 0.015, 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 }, "MiniMaxAI/MiniMax-M2.5": { input: 0.3, output: 1.2, cacheRead: 0.03, cacheWrite: 0 }, @@ -357,4 +359,6 @@ export const MODEL_COSTS: Readonly> = { }, } +// No dated promotions are active. Add an entry here when the pricing page +// publishes a promotion end date, so tests fail when it needs review. export const TEMPORARY_PRICING: readonly TemporaryPricing[] = [] diff --git a/tests/fixtures/commandcode-model-ids.json b/tests/fixtures/commandcode-model-ids.json index 0f9b907..d1d8dd0 100644 --- a/tests/fixtures/commandcode-model-ids.json +++ b/tests/fixtures/commandcode-model-ids.json @@ -1,5 +1,5 @@ { - "fetchedAt": "2026-10-16T17:58:44.197Z", + "fetchedAt": "2026-10-22T20:38:00.785Z", "source": "https://api.commandcode.ai/provider/v1/models", "modelIds": [ "claude-sonnet-5-5", @@ -42,9 +42,11 @@ "zai-org/GLM-5.2-Fast", "zai-org/GLM-5.1", "zai-org/GLM-5", + "minimax/minimax-m3.1-flash", "MiniMaxAI/MiniMax-M3", "MiniMaxAI/MiniMax-M2.7", "MiniMaxAI/MiniMax-M2.5", + "ornith/ornith-1.5-122b", "xiaomi/mimo-v2.6-pro", "xiaomi/mimo-v2.6-pro-ultraspeed", "xiaomi/mimo-v2.6-flash", @@ -80,7 +82,7 @@ "thinkingmachines/inkling", "thinkingmachines/inkling-small", "stealth/glyph-cluster:free", - "stealth/cat-jam:free", + "reflection/beam:free", "poolside/laguna-s-2.1-free", "inclusionai/ling-3.1-flash", "meta/muse-spark-1.1", diff --git a/tests/fixtures/commandcode-pricing.json b/tests/fixtures/commandcode-pricing.json index 1cecfda..0b5a75b 100644 --- a/tests/fixtures/commandcode-pricing.json +++ b/tests/fixtures/commandcode-pricing.json @@ -1,5 +1,5 @@ { - "verifiedAt": "2026-10-15", + "verifiedAt": "2026-10-22", "source": "https://commandcode.ai/docs/resources/pricing-limits", "tierPolicy": "Use request-wide input tiers; the highest threshold exceeded by input plus cache tokens applies to the full request.", "tiers": { @@ -21,9 +21,10 @@ "xai/grok-4.7": [[200000, 4, 12, 1, 0]] }, "costs": { - "stealth/cat-jam:free": [0, 0, 0, 0], "poolside/laguna-s-2.1-free": [0, 0, 0, 0], "stealth/glyph-cluster:free": [0, 0, 0, 0], + "reflection/beam:free": [0, 0, 0, 0], + "ornith/ornith-1.5-122b": [0.16, 0.64, 0.02, 0], "inclusionai/ling-3.1-flash": [0.3, 0.9, 0.06, 0], "mistral/mistral-large-4": [1.36, 4.18, 0.14, 0], "tencent/hy3-paid": [0.14, 0.58, 0.035, 0], @@ -40,6 +41,7 @@ "zai-org/GLM-5.2-Fast": [3, 10.25, 0.5, 0], "zai-org/GLM-5.1": [1.4, 4.4, 0.26, 0], "zai-org/GLM-5": [1, 3.2, 0.2, 0], + "minimax/minimax-m3.1-flash": [0.15, 0.6, 0.015, 0], "MiniMaxAI/MiniMax-M3": [0.3, 1.2, 0.06, 0], "MiniMaxAI/MiniMax-M2.7": [0.3, 1.2, 0.06, 0], "MiniMaxAI/MiniMax-M2.5": [0.3, 1.2, 0.03, 0], diff --git a/tests/test-models.ts b/tests/test-models.ts index b1e1390..90245fd 100644 --- a/tests/test-models.ts +++ b/tests/test-models.ts @@ -210,6 +210,36 @@ describe("commandCodeModelsFromApiResponse()", () => { assert.equal(MODEL_MAX_OUTPUT_TOKENS[retired], undefined) }) + it("advertises MiniMax Flash, Beam, and Ornith capabilities from the refreshed CLI", () => { + const efforts = ["low", "medium", "high", "xhigh", "max"] + for (const id of ["minimax/minimax-m3.1-flash"]) { + assert.deepEqual(inputModalitiesForModel(id), ["text", "image"]) + assert.equal(MODEL_REASONING[id], true) + assert.deepEqual(thinkingMetadataForModel(id)?.thinking?.efforts, efforts) + } + const beam = "reflection/beam:free" + assert.deepEqual(inputModalitiesForModel(beam), ["text"]) + assert.equal(MODEL_REASONING[beam], true) + assert.deepEqual(thinkingMetadataForModel(beam)?.thinking?.efforts, efforts) + assert.equal(MODEL_MAX_OUTPUT_TOKENS[beam], 131_072) + for (const id of ["ornith/ornith-1.5-122b"]) { + assert.deepEqual(inputModalitiesForModel(id), ["text", "image"]) + assert.equal(MODEL_REASONING[id], true) + assert.equal(MODEL_EFFORTS[id], undefined) + } + for (const retired of [ + "stealth/cat-jam:free", + "minimax/minimax-m3.1-flash:free", + "ornith/ornith-1.5-122b:free", + ]) { + assert.equal(MODEL_INPUT_MODALITIES[retired], undefined) + assert.equal(MODEL_REASONING[retired], undefined) + assert.equal(MODEL_EFFORTS[retired], undefined) + assert.equal(MODEL_MAX_OUTPUT_TOKENS[retired], undefined) + assert.deepEqual(inputModalitiesForModel(retired), ["text"]) + } + }) + it("prefers host-resolved input modalities over the catalog snapshot", () => { // A model published upstream after the pinned CLI release is absent from the // generated catalog, so the host's resolved modalities must win. diff --git a/tests/test-pricing-check.ts b/tests/test-pricing-check.ts index b611603..930903f 100644 --- a/tests/test-pricing-check.ts +++ b/tests/test-pricing-check.ts @@ -345,6 +345,32 @@ describe("pricing page id matching", () => { ]) }) + it("matches a provider-prefixed :free model and preserves its capacity-limited deal", () => { + const modelId = "reflection/beam:free" + assert.deepEqual(pricingPageIdCandidates(modelId), [modelId, "beam:free"]) + const rows = parsePricingPage( + pageWithRows([ + { + id: "beam:free", + tiers: [{ rates: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 } }], + deal: { + id: "beam-free", + discountPercent: 100, + free: true, + endsWhen: "while capacity lasts", + }, + }, + ]), + ) + const result = checkCommandCodePricing([modelId], rows, { [modelId]: cost(0, 0, 0, 0) }, AT_MS) + assert.deepEqual(result.issues, []) + assert.equal(result.deals.length, 1) + assert.equal(result.deals[0]!.modelId, modelId) + assert.equal(result.deals[0]!.deal.id, "beam-free") + assert.equal(result.deals[0]!.deal.endsWhen, "while capacity lasts") + assert.equal(result.deals[0]!.deal.expires, undefined) + }) + it("flags ambiguous page rows instead of picking the first match", () => { const pageRows = [pageRow("vendor/model", cost(1, 1, 0, 0)), pageRow("model", cost(1, 1, 0, 0))] const result = checkCommandCodePricing( diff --git a/tests/test-pricing.ts b/tests/test-pricing.ts index cf0b472..37007d6 100644 --- a/tests/test-pricing.ts +++ b/tests/test-pricing.ts @@ -30,7 +30,7 @@ const pricingFixture = JSON.parse(await readFile(pricingFixtureUrl, "utf-8")) as const freeModels = new Set([ "poolside/laguna-s-2.1-free", "stealth/glyph-cluster:free", - "stealth/cat-jam:free", + "reflection/beam:free", ]) function assertCost( @@ -54,7 +54,8 @@ function assertCost( describe("MODEL_COSTS pricing overlay", () => { it("covers the current Command Code model catalog snapshot", () => { assert.equal(fixture.source, "https://api.commandcode.ai/provider/v1/models") - assert.match(fixture.fetchedAt, /^2026-10-16T/) + assert.match(fixture.fetchedAt, /^\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d{3}Z$/) + assert.equal(new Date(fixture.fetchedAt).toISOString(), fixture.fetchedAt) const catalogIds = [...fixture.modelIds].sort() const pricedIds = Object.keys(MODEL_COSTS).sort() @@ -95,6 +96,29 @@ describe("MODEL_COSTS pricing overlay", () => { assert.deepEqual(MODEL_COSTS, expected) }) + it("matches reviewed MiniMax M3.1 Flash and Ornith rates after their free promotions ended", () => { + assertCost("minimax/minimax-m3.1-flash", { + input: 0.15, + output: 0.6, + cacheRead: 0.015, + cacheWrite: 0, + }) + assertCost("ornith/ornith-1.5-122b", { + input: 0.16, + output: 0.64, + cacheRead: 0.02, + cacheWrite: 0, + }) + // Both free promotions ended on 2026-10-22 and the API dropped the models. + assert.equal(MODEL_COSTS["minimax/minimax-m3.1-flash:free"], undefined) + assert.equal(MODEL_COSTS["ornith/ornith-1.5-122b:free"], undefined) + assert.deepEqual(TEMPORARY_PRICING, []) + }) + + it("matches reviewed late-October Beam promotion rates", () => { + assertCost("reflection/beam:free", { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }) + }) + it("uses non-zero prices except for models documented as free", () => { for (const [modelId, cost] of Object.entries(MODEL_COSTS)) { assert.ok(cost.input >= 0, `${modelId} input cost should be non-negative`) @@ -368,12 +392,12 @@ describe("MODEL_COSTS pricing overlay", () => { cacheRead: 0.15, cacheWrite: 0, }) - assertCost("stealth/cat-jam:free", { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 }) + assert.equal(MODEL_COSTS["stealth/cat-jam:free"], undefined) }) it("tracks pricing provenance", () => { assert.equal(PRICING_SOURCE_URL, "https://commandcode.ai/docs/resources/pricing-limits") - assert.equal(PRICING_LAST_VERIFIED, "2026-10-15") + assert.equal(PRICING_LAST_VERIFIED, "2026-10-22") }) it("fails once temporary pricing needs review", () => { @@ -383,7 +407,7 @@ describe("MODEL_COSTS pricing overlay", () => { assert.ok(pricing.models.length > 0) assert.ok( pricing.expiresOn >= today, - `${pricing.description} for ${pricing.models.join(", ")} expired on ${pricing.expiresOn}; refresh MODEL_COSTS`, + `${pricing.description} for ${pricing.models.join(", ")} needs recheck after ${pricing.expiresOn}; refresh MODEL_COSTS`, ) for (const modelId of pricing.models) { assert.ok(MODEL_COSTS[modelId], `${modelId} should have a temporary price entry`)