chore: sync models with dashboard API

README.md

@@ -141,9 +141,6 @@ zerogpu chat "Explique la mise en cache en une phrase." -m qwen3-30b-a3b-fp8
 
 # A 262K-token context, for whole repos and very long documents
 zerogpu chat "$(cat ARCHITECTURE.md)" -m glm-5.2
-
-# Coding and agentic work, at a fraction of the flagship price
-zerogpu chat "Port this helper to async/await." -m deepseek-v4-flash-0731
 ```
 
 | Option | Description |
@@ -159,11 +156,14 @@ zerogpu chat "Port this helper to async/await." -m deepseek-v4-flash-0731
 | `gpt-oss-120b` | 120B MoE, 131K context, reasoning + function calling. |
 | `llama-guard-4-12b` | 12B dense, 164K context, brand safety + text moderation. |
 | `deepseek-v4.1-flash` | Sparse MoE (8B active on input, 16B on output), 1M context, long-context agentic work + function calling. |
+| `glm-5.3-flash` | Hybrid sparse + linear attention, 1M context, coding and long-horizon agentic work + function calling. |
+| `gpt-5.6-luna` | Cost-optimized GPT-5.6, 272K context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
+| `gpt-4.1-mini` | Fast, cost-efficient GPT-4.1, 1M context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
+| `gpt-5.4-nano` | Most cost-efficient GPT-5.4, 400K context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
 | `qwen3-30b-a3b-fp8` | 30B MoE, 100+ languages, reasoning + function calling. |
 | `glm-5.2` | 753B MoE, 262K context, reasoning + function calling. The platform's most capable model, and its priciest. |
-| `deepseek-v4-flash-0731` | 284B MoE (13B active), 1M context, coding and agentic workflows. |
 
-`qwen3-30b-a3b-fp8`, `glm-5.2`, and `deepseek-v4-flash-0731` are served by the Chat Completions API rather than the Responses API; the CLI routes them automatically.
+`qwen3-30b-a3b-fp8` and `glm-5.2` are served by the Chat Completions API rather than the Responses API; the CLI routes them automatically.
 
 #### `chat_thinking`
 

docs/ADDING_COMMANDS.md

@@ -9,7 +9,7 @@ This guide explains how to add a new CLI command to the ZeroGPU CLI.
 - `src/commands/` — one file per command, each exporting a `register<Name>Command(program)` function.
 - `src/cli.ts` — wires every command into the root program.
 - `src/lib/responses.ts` — shared `RESPONSES_ENDPOINT`, `ResponsesApiResponse`, and the `extractOutputText` / `extractReasoningText` helpers for `/v1/responses` calls.
-- `src/lib/chatCompletions.ts` — the same for `/v1/chat/completions`, used by models the platform serves only there (currently `qwen3-30b-a3b-fp8`, `glm-5.2`, and `deepseek-v4-flash-0731`), plus `toResponsesUsage` to normalize token counts for savings tracking.
+- `src/lib/chatCompletions.ts` — the same for `/v1/chat/completions`, used by models the platform serves only there (currently `qwen3-30b-a3b-fp8` and `glm-5.2`), plus `toResponsesUsage` to normalize token counts for savings tracking.
 - `src/lib/auth.ts` — `getApiKey()` for authenticated requests.
 - `src/lib/request.ts` — plumbing for the endpoint commands: `requireApiKey`, `resolveInput` (argument or stdin), `parseJsonObject`, `postJson`, and the print helpers.
 

docs/DOCUMENTATION.md

@@ -5,7 +5,7 @@
 `zerogpu-cli` is the official command-line interface for [ZeroGPU](https://zerogpu.ai), a distributed / edge inference platform for small language models (SLMs) and nano language models. The CLI is a thin, OpenAI-compatible client around the ZeroGPU **Responses API** (`https://api.zerogpu.ai/v1/responses`) — and, for models served only there, the **Chat Completions API** (`https://api.zerogpu.ai/v1/chat/completions`) — that lets you call a curated set of edge-optimized models directly from your terminal for common NLP workloads:
 
 - Conversational chat (`LFM2.5-1.2B-Instruct`, `LFM2.5-1.2B-Thinking`)
-- Reasoning and tool-use chat (`gpt-oss-120b`, `llama-guard-4-12b`, `deepseek-v4.1-flash`, `qwen3-30b-a3b-fp8`, `glm-5.2`, `deepseek-v4-flash-0731`)
+- Reasoning and tool-use chat (`gpt-oss-120b`, `llama-guard-4-12b`, `deepseek-v4.1-flash`, `glm-5.3-flash`, `gpt-5.6-luna`, `gpt-4.1-mini`, `gpt-5.4-nano`, `qwen3-30b-a3b-fp8`, `glm-5.2`)
 - IAB content/audience classification (`zlm-v1-iab-classify-edge`, `zlm-v2-iab-classify-edge-enriched`)
 - Domain-level IAB classification (`zlm-v1-iab-domain-classifier`)
 - Zero-shot classification (`deberta-v3-small`)
@@ -214,11 +214,14 @@ zerogpu chat <text> [-i <instructions>] [-m <model>] [-r]
 | `gpt-oss-120b` | Responses | 120B MoE, 131K context, reasoning + function calling. |
 | `llama-guard-4-12b` | Responses | 12B dense, 163,840-token context, brand safety + text moderation. |
 | `deepseek-v4.1-flash` | Responses | Sparse MoE (8B active on input, 16B on output), 1,048,576-token context, long-context agentic work + function calling. |
+| `glm-5.3-flash` | Responses | Hybrid sparse + linear attention, 1,048,576-token context, coding and long-horizon agentic work + function calling. |
+| `gpt-5.6-luna` | Responses | Cost-optimized GPT-5.6, 272,000-token context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
+| `gpt-4.1-mini` | Responses | Fast, cost-efficient GPT-4.1, 1,047,576-token context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
+| `gpt-5.4-nano` | Responses | Most cost-efficient GPT-5.4, 400,000-token context, coding, chat, function calling, reasoning, RAG, summarization + translation. |
 | `qwen3-30b-a3b-fp8` | Chat Completions | 30B MoE, 100+ languages, reasoning + function calling. |
 | `glm-5.2` | Chat Completions | 753B MoE, 262,144-token context, reasoning + function calling. The most capable model on the platform, and the most expensive by an order of magnitude. |
-| `deepseek-v4-flash-0731` | Chat Completions | 284B MoE (13B active), 1,048,576-token context, coding and agentic workflows. |
 
-`qwen3-30b-a3b-fp8`, `glm-5.2`, and `deepseek-v4-flash-0731` have no Responses endpoint, so the CLI posts them to `/v1/chat/completions` instead, mapping `--instructions` to a `system` message and normalizing `prompt_tokens` / `completion_tokens` back to Responses token names for savings tracking. This routing is transparent — the command and its output are identical either way.
+`qwen3-30b-a3b-fp8` and `glm-5.2` have no Responses endpoint, so the CLI posts them to `/v1/chat/completions` instead, mapping `--instructions` to a `system` message and normalizing `prompt_tokens` / `completion_tokens` back to Responses token names for savings tracking. This routing is transparent — the command and its output are identical either way.
 
 **Example**
 ```bash
@@ -815,7 +818,7 @@ Content-Type: application/json
 x-api-key:    <ZEROGPU_API_KEY>
 ```
 
-The exceptions are `summarize`, `chat --model qwen3-30b-a3b-fp8`, `chat --model glm-5.2`, and `chat --model deepseek-v4-flash-0731`, whose models the ZeroGPU platform serves only through the OpenAI-compatible Chat Completions endpoint:
+The exceptions are `summarize`, `chat --model qwen3-30b-a3b-fp8`, and `chat --model glm-5.2`, whose models the ZeroGPU platform serves only through the OpenAI-compatible Chat Completions endpoint:
 
 ```
 POST https://api.zerogpu.ai/v1/chat/completions

package.json

@@ -1,6 +1,6 @@
 {
   "name": "zerogpu-cli",
-  "version": "3.10.1",
+  "version": "3.11.0",
   "description": "Command-line interface for ZeroGPU.",
   "type": "module",
   "bin": {

src/commands/chat.ts

@@ -17,18 +17,21 @@ import { recordAndMaybeNotify } from "../lib/savings.js";
 const DEFAULT_MODEL = "LFM2.5-1.2B-Instruct";
 
 // Text-generation models `--model` accepts, and the API each one speaks.
-// qwen3-30b-a3b-fp8, glm-5.2, and deepseek-v4-flash-0731 are Chat Completions
-// only — they have no Responses endpoint.
+// qwen3-30b-a3b-fp8 and glm-5.2 are Chat Completions only — they have no
+// Responses endpoint.
 // Source: https://docs.zerogpu.ai/docs/text-generation
 const CHAT_MODELS: Record<string, "responses" | "chat-completions"> = {
   "LFM2.5-1.2B-Instruct": "responses",
   "LFM2.5-1.2B-Thinking": "responses",
   "gpt-oss-120b": "responses",
   "llama-guard-4-12b": "responses",
   "deepseek-v4.1-flash": "responses",
+  "glm-5.3-flash": "responses",
+  "gpt-5.6-luna": "responses",
+  "gpt-4.1-mini": "responses",
+  "gpt-5.4-nano": "responses",
   "qwen3-30b-a3b-fp8": "chat-completions",
   "glm-5.2": "chat-completions",
-  "deepseek-v4-flash-0731": "chat-completions",
 };
 
 // Model ids are case-sensitive to the API but not to the person typing them.

src/lib/savings.ts

@@ -21,12 +21,15 @@ const DEFAULT_BASELINE = "claude-opus-4-8";
 // this table to the published catalog — see tests/savings.test.ts.
 export const ZGPU_PRICING: Record<string, { in: number; out: number }> = {
   "gpt-oss-120b": { in: 0.15, out: 0.6 },
-  "qwen3-30b-a3b-fp8": { in: 0.05, out: 0.3 },
+  "qwen3-30b-a3b-fp8": { in: 0.1, out: 0.45 },
   "glm-5.2": { in: 1.1, out: 3.5 },
-  "deepseek-v4.1-flash": { in: 0.3, out: 1.2 },
-  "deepseek-v4-flash-0731": { in: 0.16, out: 0.38 },
+  "deepseek-v4.1-flash": { in: 0.14, out: 0.57 },
+  "glm-5.3-flash": { in: 0.1, out: 0.35 },
+  "gpt-5.6-luna": { in: 0.2, out: 1.2 },
+  "gpt-4.1-mini": { in: 0.4, out: 1.6 },
+  "gpt-5.4-nano": { in: 0.2, out: 1.25 },
   "llama-guard-4-12b": { in: 0.18, out: 0.18 },
-  "llama-3.1-8b-instruct-fast": { in: 0.02, out: 0.05 },
+  "llama-3.1-8b-instruct-fast": { in: 0.15, out: 0.28 },
   "zlm-v1-iab-classify-edge": { in: 0.02, out: 0.05 },
   "zlm-v2-iab-classify-edge-enriched": { in: 0.025, out: 0.15 },
   "zlm-v1-iab-domain-classifier": { in: 0.02, out: 0.05 },
@@ -37,6 +40,9 @@ export const ZGPU_PRICING: Record<string, { in: number; out: number }> = {
   "LFM2.5-1.2B-Thinking": { in: 0.02, out: 0.05 },
   "LFM2.5-1.2B-Instruct": { in: 0.02, out: 0.05 },
   "zlm-v1-moderation-edge": { in: 0.02, out: 0.05 },
+  // Audio models: the catalog publishes $0 in and $0 out for both.
+  "whisper-tiny": { in: 0, out: 0 },
+  "chatterbox-nano": { in: 0, out: 0 },
   // Embedding models bill input tokens only; there are no output tokens to
   // charge, so `out: 0` is the real rate, not a placeholder.
   "all-minilm-l6-v2": { in: 0.004, out: 0 },

tests/savings.test.ts

@@ -52,7 +52,7 @@ describe("computeCallSavings", () => {
     );
     expect(tokens).toBe(1020);
     const claude = (740 * 5 + 280 * 25) / 1e6; // 0.0107
-    const zgpu = (740 * 0.02 + 280 * 0.05) / 1e6; // real ZeroGPU cost
+    const zgpu = (740 * 0.15 + 280 * 0.28) / 1e6; // real ZeroGPU cost
     expect(savingsUsd).toBeCloseTo(claude - zgpu, 8);
   });
 
@@ -93,7 +93,6 @@ describe("computeCallSavings", () => {
       "gpt-oss-120b",
       "qwen3-30b-a3b-fp8",
       "glm-5.2",
-      "deepseek-v4-flash-0731",
       "LFM2.5-1.2B-Instruct",
     ]) {
       expect(unknown).toBeLessThanOrEqual(
@@ -112,12 +111,15 @@ describe("ZGPU_PRICING tracks the published model catalog", () => {
   // gpt-oss-120b sat at $0.03/$0.10 long after it was repriced to $0.15/$0.60.
   const CATALOG: Record<string, { in: number; out: number }> = {
     "gpt-oss-120b": { in: 0.15, out: 0.6 },
-    "qwen3-30b-a3b-fp8": { in: 0.05, out: 0.3 },
+    "qwen3-30b-a3b-fp8": { in: 0.1, out: 0.45 },
     "glm-5.2": { in: 1.1, out: 3.5 },
-    "deepseek-v4.1-flash": { in: 0.3, out: 1.2 },
-    "deepseek-v4-flash-0731": { in: 0.16, out: 0.38 },
+    "deepseek-v4.1-flash": { in: 0.14, out: 0.57 },
+    "glm-5.3-flash": { in: 0.1, out: 0.35 },
+    "gpt-5.6-luna": { in: 0.2, out: 1.2 },
+    "gpt-4.1-mini": { in: 0.4, out: 1.6 },
+    "gpt-5.4-nano": { in: 0.2, out: 1.25 },
     "llama-guard-4-12b": { in: 0.18, out: 0.18 },
-    "llama-3.1-8b-instruct-fast": { in: 0.02, out: 0.05 },
+    "llama-3.1-8b-instruct-fast": { in: 0.15, out: 0.28 },
     "zlm-v2-iab-classify-edge-enriched": { in: 0.025, out: 0.15 },
     "zlm-v1-iab-classify-edge": { in: 0.02, out: 0.05 },
     "zlm-v1-iab-domain-classifier": { in: 0.02, out: 0.05 },
@@ -128,6 +130,9 @@ describe("ZGPU_PRICING tracks the published model catalog", () => {
     "LFM2.5-1.2B-Thinking": { in: 0.02, out: 0.05 },
     "LFM2.5-1.2B-Instruct": { in: 0.02, out: 0.05 },
     "zlm-v1-moderation-edge": { in: 0.02, out: 0.05 },
+    // Audio models: the catalog publishes $0 in and $0 out for both.
+    "whisper-tiny": { in: 0, out: 0 },
+    "chatterbox-nano": { in: 0, out: 0 },
     "all-minilm-l6-v2": { in: 0.004, out: 0 },
     "bge-small-en-v1.5": { in: 0.004, out: 0 },
   };