From 08492a061347706aaf09226ca8271194d33cf004 Mon Sep 17 00:00:00 2001 From: Claude Date: Thu, 29 Oct 2026 14:09:00 +0000 Subject: [PATCH] chore: sync models with dashboard API MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - qwen3-30b-a3b-fp8: input $0.05 -> $0.10, output $0.30 -> $0.45 (ZGPU_PRICING, test CATALOG) - llama-3.1-8b-instruct-fast: input $0.02 -> $0.15, output $0.05 -> $0.28 (ZGPU_PRICING, test CATALOG, sample-call test) - zlm-v2-iab-classify-edge-enriched: input $0.025 -> $0.05 (ZGPU_PRICING, test CATALOG) - zlm-v1-signal-extract: input $0.02 -> $0.04, output $0.05 -> $0.10 (ZGPU_PRICING, test CATALOG) - zlm-v1-moderation-edge: input $0.02 -> $0.10, output $0.05 -> $0.15 (ZGPU_PRICING, test CATALOG) - ZGPU_FALLBACK: $1.10/$3.50 -> $0.40/$1.60, now gpt-4.1-mini (comment, fallback test) - add glm-5.3-flash, gpt-4.1-mini, gpt-6-luna, gpt-5.6-luna, gpt-5.4-nano, qwen3.6-35b-a3b, zlm-v1-tool-decisions: pricing, chat --model (Responses), README + DOCUMENTATION rows - add llama-3.2-3b-instruct-fast, gliner2.5-base-v1, gliner2.5-multi-v1, gliner2.5-decide-v1, whisper-tiny, chatterbox-nano: pricing only — callable through the endpoint commands - remove glm-5.2, deepseek-v4-flash-0731: pricing, chat --model, docs, examples, routing prose - remove gliner2-base-v1: pricing, docs; commands classify_structured, extract_entities, extract_json deleted - remove gliner-multi-pii-v1: pricing, docs; commands extract_pii, redact_pii deleted - DOCUMENTATION sections renumbered 4.8-4.13 after the five deletions - version 3.10.1 -> 3.11.0 Source: https://api-dashboard.zerogpu.ai/api/models Co-Authored-By: Claude Opus 5 Claude-Session: https://claude.ai/code/session_01V6BqP2DJdbxt3DayLfo7r7 --- README.md | 94 ++--------- docs/ADDING_COMMANDS.md | 2 +- docs/DOCUMENTATION.md | 262 ++++------------------------- package-lock.json | 4 +- package.json | 2 +- src/cli.ts | 10 -- src/commands/chat.ts | 12 +- src/commands/classifyStructured.ts | 90 ---------- src/commands/extractEntities.ts | 122 -------------- src/commands/extractJson.ts | 90 ---------- src/commands/extractPii.ts | 105 ------------ src/commands/redactPii.ts | 77 --------- src/lib/savings.ts | 32 ++-- tests/endpointCommands.test.ts | 12 +- tests/savings.test.ts | 51 +++--- 15 files changed, 109 insertions(+), 856 deletions(-) delete mode 100644 src/commands/classifyStructured.ts delete mode 100644 src/commands/extractEntities.ts delete mode 100644 src/commands/extractJson.ts delete mode 100644 src/commands/extractPii.ts delete mode 100644 src/commands/redactPii.ts diff --git a/README.md b/README.md index eca8232..2e5e54e 100644 --- a/README.md +++ b/README.md @@ -25,13 +25,7 @@ The official command-line interface for [ZeroGPU](https://zerogpu.ai) — run fa - [`classify_iab`](#classify_iab) - [`classify_iab_enriched`](#classify_iab_enriched) - [`classify_domain`](#classify_domain) - - [`classify_structured`](#classify_structured) - [`classify_zero_shot`](#classify_zero_shot) - - [Extraction](#extraction) - - [`extract_entities`](#extract_entities) - - [`extract_json`](#extract_json) - - [`extract_pii`](#extract_pii) - - [`redact_pii`](#redact_pii) - [Endpoints](#endpoints) - [`responses`](#responses) - [`chat_completions`](#chat_completions) @@ -138,12 +132,6 @@ zerogpu chat "Why does my API keep getting rate-limited?" -m gpt-oss-120b -r # Multilingual reasoning 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 +147,16 @@ 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` | 1M context, coding and long-horizon agent work + function calling. | +| `gpt-4.1-mini` | 1M context, coding + function calling, with structured outputs. | +| `gpt-6-luna` | 272K context, reasoning + function calling, text and image input. | +| `gpt-5.6-luna` | 272K context, cost-optimized reasoning + function calling. | +| `gpt-5.4-nano` | 400K context, classification, extraction, and routing + function calling. | +| `qwen3.6-35b-a3b` | 35B MoE, 16K context, multimodal chat — document, image, and video analysis + data extraction. | +| `zlm-v1-tool-decisions` | 0.6B, 24K context, MCP tool selection and argument generation for agents. | | `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` is served by the Chat Completions API rather than the Responses API; the CLI routes it automatically. #### `chat_thinking` @@ -211,19 +204,6 @@ zerogpu classify_domain nytimes.com Returns scored content categories plus topics, keywords, and inferred user intent for the domain. When you have the page text and need per-URL precision, use `classify_iab` instead. -#### `classify_structured` - -Classify text against a **structured schema** of categories and allowed labels using GLiNER2. - -```bash -zerogpu classify_structured "The product arrived broken and I want a refund." \ - --schema '{"sentiment":["positive","negative","neutral"],"intent":["complaint","praise","question"]}' -``` - -| Option | Description | -|---|---| -| `-s, --schema ` | **Required.** JSON object mapping category name → allowed labels. | - #### `classify_zero_shot` Zero-shot classification against arbitrary candidate labels using **DeBERTa v3**. @@ -245,60 +225,6 @@ zerogpu classify_zero_shot "The stock market dropped 3% today." \ --- -### Extraction - -#### `extract_entities` - -Extract named entities from text using **GLiNER2** with custom labels. - -```bash -zerogpu extract_entities "Tim Cook visited the Cupertino campus on Tuesday." \ - --label person --label organization --label location --threshold 0.4 -``` - -| Option | Default | Description | -|---|---|---| -| `-l, --label