@oonacode/linux-x64 0.3.244 → 0.3.246

package/oona

@@ -1,4 +1,4 @@
 [diffend] Oversized file quarantined before diffing.
 name: package/oona
-size: 107300320 bytes
-sha256: 93e3b37818b70c326b469b776106af863b83b8746c72e31c2b0b036c4646ab35
+size: 107386336 bytes
+sha256: 7ab938c45e880c2f99fde4ea31724dbac79d23eb8ecbeb00abf88a54750dad85

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@oonacode/linux-x64",
-  "version": "0.3.244",
+  "version": "0.3.246",
   "description": "OonaCode native binaries for Linux x64: oona (CLI + headless host service), the pinned oonaclaude engine sidecar, local web UI assets. Installed automatically as an optional dependency of `oonacode`.",
   "license": "UNLICENSED",
   "homepage": "https://oonacode.oonak.ai",

package/registry.json

@@ -1210,7 +1210,18 @@
           "supportsTools": true,
           "supportsThinking": true,
           "supportsVision": false,
-          "notes": "DeepSeek's current Flash model (its API calls it `deepseek-flash`, the docs name it DeepSeek-V4.1-Flash; the legacy `deepseek-v4-flash` ids still answer and now cost the same). Listed under the canonical id Alibaba and everyone else use so the pool can fold the two backends into one entry; `upstreamId` is what DeepSeek is sent. Peak price in the row ($0.30 in / $1.20 out, cache hit $0.006); off-peak is half, outside 01:00-04:00 and 06:00-10:00 UTC Mon-Fri. Owner, 2026-09-18: \"add deepseek 4.1 in oonacode pool\"."
+          "notes": "DeepSeek's current Flash model (its API calls it `deepseek-flash`, which has been the 1017 snapshot since 2026-10-17 - see `deepseek-v4.1-flash-1017` below; the docs name it DeepSeek-V4.1-Flash; the legacy `deepseek-v4-flash` ids still answer and now cost the same). Listed under the canonical id Alibaba and everyone else use so the pool can fold the two backends into one entry; `upstreamId` is what DeepSeek is sent. Peak price in the row ($0.30 in / $1.20 out, cache hit $0.006); off-peak is half, outside 01:00-04:00 and 06:00-10:00 UTC Mon-Fri. Owner, 2026-09-18: \"add deepseek 4.1 in oonacode pool\"."
+        },
+        {
+          "id": "deepseek-v4.1-flash-1017",
+          "displayName": "DeepSeek V4.1 Flash 1017",
+          "upstreamId": "deepseek-flash",
+          "contextWindow": 1000000,
+          "maxOutputTokens": 384000,
+          "supportsTools": true,
+          "supportsThinking": true,
+          "supportsVision": true,
+          "notes": "DeepSeek-V4.1-Flash-1017, the 17 Oct 2026 snapshot of V4.1 Flash: stronger reasoning, coding, agent and image scores, shorter outputs. DeepSeek's API serves it under the same name as the model above (`deepseek-flash`; docs api-docs.deepseek.com/quick_start/pricing: the legacy `deepseek-v4-flash` and `deepseek-v4-flash-vision-exp` ids are served by it too), so this entry is the explicit name of what that API name now is. 1M context, 384K output, image input, thinking and non-thinking, same price as V4.1 Flash: peak $0.30 in / $1.20 out, cache hit $0.006, off-peak half (peak 01:00-04:00 and 06:00-10:00 UTC Mon-Fri). Owner, 2026-10-19: \"add this model deepseek-v4.1-flash-1017 in oonacode pool\"."
         },
         {
           "id": "deepseek-v4-pro",
@@ -1264,91 +1275,6 @@
       "enabled": true,
       "docsUrl": "https://api-docs.deepseek.com/guides/anthropic_api"
     },
-    {
-      "id": "glm",
-      "displayName": "GLM (Zhipu / Z.ai)",
-      "kind": "anthropic-passthrough",
-      "route": "gateway",
-      "credentialClass": "api_key",
-      "baseUrl": "https://api.z.ai/api/anthropic",
-      "messagesPath": "/v1/messages",
-      "auth": {
-        "header": "authorization-bearer"
-      },
-      "models": [
-        {
-          "id": "glm-5.3",
-          "displayName": "GLM-5.3",
-          "contextWindow": 1000000,
-          "maxOutputTokens": 128000,
-          "supportsTools": true,
-          "supportsThinking": true,
-          "supportsVision": false,
-          "inputPricePerM": 1.4,
-          "outputPricePerM": 4.4,
-          "notes": "Flagship, text-only, reasoning always on. Verified against the Coding Plan on 2026-09-03: the API answers and reports model=glm-5.3."
-        },
-        {
-          "id": "glm-5.3-flash",
-          "displayName": "GLM-5.3 Flash",
-          "contextWindow": 1000000,
-          "maxOutputTokens": 128000,
-          "supportsTools": true,
-          "supportsThinking": true,
-          "supportsVision": true,
-          "inputPricePerM": 0.15,
-          "outputPricePerM": 0.5,
-          "notes": "320B/18B MoE, natively multimodal, 3x the Coding Plan quota of GLM-5.3. Priced at the UNDISCOUNTED rate: z.ai's listed $0.075/$0.25 is a 50% promotion ending 2026-09-09."
-        }
-      ],
-      "defaultModel": "glm-5.3",
-      "modelAliases": {
-        "opus": "glm-5.3",
-        "sonnet": "glm-5.3",
-        "haiku": "glm-5.3-flash"
-      },
-      "capabilities": {
-        "streaming": true,
-        "tools": true,
-        "thinking": true,
-        "vision": false,
-        "countTokens": false,
-        "promptCaching": true
-      },
-      "passthrough": {
-        "stripCacheControl": false,
-        "allowedBetas": [],
-        "anthropicVersion": "2023-06-01"
-      },
-      "engineEnv": {
-        "API_TIMEOUT_MS": "3000000"
-      },
-      "directCapable": true,
-      "phase": 1,
-      "enabled": true,
-      "docsUrl": "https://docs.z.ai/devpack/tool/claude",
-      "references": [
-        {
-          "label": "Claude Code integration (GLM Coding Plan)",
-          "url": "https://docs.z.ai/devpack/tool/claude",
-          "kind": "docs"
-        }
-      ],
-      "notes": "Z.ai GLM Coding Plan speaks the Anthropic Messages API natively. Per the Z.ai Claude Code guide the opus/sonnet aliases map to GLM-4.7 and the engine needs a long API_TIMEOUT_MS (3000000). Plan tiers (Lite/Pro/Max) are concurrency-limited, not token-limited.",
-      "credentialVariants": [
-        {
-          "id": "payg",
-          "label": "Pay-as-you-go (API key)",
-          "baseUrl": "https://api.z.ai/api/anthropic"
-        },
-        {
-          "id": "token-plan",
-          "label": "GLM Coding Plan (subscription)",
-          "baseUrl": "https://api.z.ai/api/anthropic",
-          "notes": "Same endpoint and same key shape as pay-as-you-go — the plan changes how the calls are billed, not where they go. Kept as its own record so its prices, limits and cache TTL are set independently."
-        }
-      ]
-    },
     {
       "id": "qwen",
       "displayName": "Alibaba Cloud (Qwen)",
@@ -1706,19 +1632,6 @@
           "outputPricePerM": 3.75,
           "notes": "Google lists one model per level — gemini-3.8-flash-high, -medium and -low (fetchAvailableModels, 2026-10-03: 1,048,576 context / 65,536 output each); OonaCode shows ONE model and the chat’s effort picker fills the level (upstreamId template; xhigh and max read as high; no effort = high). Cost = Google’s Gemini API list price for gemini-3.8-flash: $0.75 in / $0.075 cached / $3.75 out per 1M through 2026-12-31, DOUBLING to $1.50 / $0.15 / $7.50 from 2027-01-01 (ai.google.dev/gemini-api/docs/pricing, updated 2026-10-01) — move the cost cells on the model page then."
         },
-        {
-          "id": "gemini-3.7-flash",
-          "displayName": "Gemini 3.7 Flash (Antigravity)",
-          "upstreamId": "gemini-3.7-flash-{effort:high}",
-          "contextWindow": 1048576,
-          "maxOutputTokens": 65536,
-          "supportsTools": true,
-          "supportsThinking": true,
-          "supportsVision": true,
-          "inputPricePerM": 0.75,
-          "outputPricePerM": 3.75,
-          "notes": "Google lists one model per level — gemini-3.7-flash-high, -medium and -low (fetchAvailableModels, 2026-10-03: 1,048,576 context / 65,536 output each); OonaCode shows ONE model and the chat’s effort picker fills the level (upstreamId template; xhigh and max read as high; no effort = high). Cost = Google’s Gemini API list price for gemini-3.7-flash: $0.75 in / $0.075 cached / $3.75 out per 1M through 2026-12-31, DOUBLING to $1.50 / $0.15 / $7.50 from 2027-01-01 (ai.google.dev/gemini-api/docs/pricing, updated 2026-10-01) — move the cost cells on the model page then."
-        },
         {
           "id": "gemini-4-argon",
           "displayName": "Gemini 4 Argon (Antigravity)",