> ## Documentation Index
> Fetch the complete documentation index at: https://docs.laozhang.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Popular AI Models
> July 2026 guide to GPT-5.6 Sol/Terra/Luna, Claude Opus 5, Fable 5, Sonnet 5, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, Grok 4.5, GPT-Image-2, Nano Banana 2/Lite, Seedance 2.0, and other popular AI models.
## Current Model Recommendations (Updated July 25, 2026)
LaoZhang API supports 200+ mainstream AI models. This page highlights current text, reasoning, image, and video choices, including GPT-5.6 Sol / Terra / Luna, Claude Opus 5 / Fable 5 / Sonnet 5, Gemini 3.6 Flash / 3.5 Flash-Lite, Grok 4.5, GPT-Image-2, Nano Banana 2 / Lite, and Seedance 2.0.
**Distinguish provider releases from account availability**
A provider announcement does not mean every LaoZhang API token group can already call that model. Confirm model IDs, token type, region, price, and billing behavior on the [live console model list](https://api2.laozhang.ai/account/pricing). Product names are not automatically valid API model IDs.
## π₯ Currently Recommended Models
Recommendations below combine provider information available through **July 25, 2026** with LaoZhang API's current integration status. For the complete model list and live pricing, visit the [model catalog and pricing page](/en/models) or [console pricing page](https://api2.laozhang.ai/account/pricing).
Context figures below use each provider's native API limit. LaoZhang API or another gateway may expose a lower limit for a particular route, region, or plan.
## Model Categories
### π€ OpenAI Series
#### GPT-5.6 Series (Latest) π₯
| Model Name | Model ID | Context | Features | Recommended Use |
| -------------------- | ------------------------- | ------- | --------------------------------------- | --------------------------------------------------------- |
| **GPT-5.6 Sol** βββ | `gpt-5.6` / `gpt-5.6-sol` | 1.05M | Current flagship; `gpt-5.6` aliases Sol | Hard coding, agents, professional work, complex workflows |
| **GPT-5.6 Terra** ββ | `gpt-5.6-terra` | 1.05M | Balanced intelligence, speed, and cost | Production agents, coding, general knowledge work |
| **GPT-5.6 Luna** β | `gpt-5.6-luna` | 1.05M | Cost-sensitive, high-throughput tier | Batch jobs, classification, extraction, automation |
`gpt-5.6` is an alias for `gpt-5.6-sol`. GPT-5.6 Pro is enabled with the request parameter `reasoning.mode: "pro"`; it is not a `gpt-5.6-pro` model ID. Check the LaoZhang API console for tier availability and Pro-mode passthrough.
#### GPT-5.5 Series (Previous-Generation Compatibility)
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------- | ------------- | ------- | ------------------------------------------------ | --------------------------------- |
| **GPT-5.5** | `gpt-5.5` | 1.05M | Previous-generation general flagship | Existing integrations |
| **GPT-5.5 Pro** | `gpt-5.5-pro` | 1.05M | Previous-generation high-compute reasoning model | Existing hard-reasoning workflows |
#### GPT-5.1 Series (November 2025) π₯
| Model Name | Model ID | Context | Features | Recommended Use |
| ---------------------- | -------------------- | ------- | ---------------------------- | ----------------------------------------- |
| **GPT-5.1** ββ | `gpt-5.1` | 400K | Strong performance, balanced | General advanced tasks |
| **GPT-5.1-Codex** ββ | `gpt-5.1-codex` | 400K | Responses API coding model | Programming and long-running coding tasks |
| **GPT-5.1-Codex Mini** | `gpt-5.1-codex-mini` | 400K | Lightweight coding model | Quick coding and completion |
The official Codex model IDs are `gpt-5.1-codex` and `gpt-5.1-codex-mini`. High and medium are reasoning-effort settings, not `gpt-5.1-codex-high` or `gpt-5.1-codex-medium` model IDs.
#### GPT-5 Series (August 2025)
| Model Name | Model ID | Context | Features | Recommended Use |
| -------------- | ------------ | ------- | ---------------------------- | ------------------------ |
| **GPT-5** | `gpt-5` | 400K | First-generation GPT-5 | General tasks |
| **GPT-5 Pro** | `gpt-5-pro` | 400K | High-compute reasoning model | Complex enterprise tasks |
| **GPT-5 Mini** | `gpt-5-mini` | 400K | Lightweight efficient | Cost-sensitive scenarios |
| **GPT-5 Nano** | `gpt-5-nano` | 400K | Ultra-lightweight | Batch processing |
#### Reasoning Models
| Model Name | Model ID | Context | Features | Recommended Use |
| -------------- | --------- | ------- | --------------------- | ------------------ |
| **o3-pro** ββ | `o3-pro` | 200K | Strongest reasoning | Top-tier reasoning |
| **o3** β | `o3` | 200K | Reasoning model | Complex reasoning |
| **o4-mini** ββ | `o4-mini` | 200K | Lightweight reasoning | Programming tasks |
#### GPT-4 Series
| Model Name | Model ID | Context | Features | Recommended Use |
| ---------------- | -------------- | ------- | -------------------------------- | ------------------------------------ |
| **GPT-4.1** β | `gpt-4.1` | 1.05M | Non-reasoning long-context model | General applications, long documents |
| **GPT-4.1 Mini** | `gpt-4.1-mini` | 1.05M | Affordable lightweight | Cost-sensitive |
| **GPT-4.1 Nano** | `gpt-4.1-nano` | 1.05M | Ultra-low-cost | High-volume |
| **GPT-4o** | `gpt-4o` | 128K | Balanced multimodal | General scenarios |
| **GPT-4o Mini** | `gpt-4o-mini` | 128K | Lightweight fast | Quick responses |
#### Image Generation
| Model Name | Model ID | Features | Recommended Use |
| ----------------------- | --------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------- |
| **GPT-Image-2** βββ | `gpt-image-2` / `gpt-image-2-vip` | Generation and editing; official API accepts dimensions divisible by 16, with a 3840px maximum edge and 8,294,400-pixel cap; VIP commonly offers 1K / 2K / 4K | OpenAI-compatible production image workflows |
| **GPT-Image-1.5** ββ | `gpt-image-1.5` | Previous high-quality generation model | Existing professional design workflows |
| **GPT-Image-1** β | `gpt-image-1` | High cost-performance | General image generation |
| **GPT-Image-1 Mini** | `gpt-image-1-mini` | Lightweight fast | Quick generation |
| **Flux Kontext Max** ββ | `flux-kontext-max` | Highest quality | Professional design |
| **Flux Kontext Pro** β | `flux-kontext-pro` | Professional quality | Commercial design |
| **DALLΒ·E 3** | `dall-e-3` | Classic generation | Standard tasks |
#### Video Generation
| Model Name | Model ID | Features | Recommended Use |
| ----------------------------------- | --------------------------------- | ------------------------------------------------------------------------ | ------------------------------------------------ |
| **Seedance 2.0 Fast** ββ | `doubao-seedance-2-0-fast-260128` | Fast multimodal generation with text, image, video, and audio references | Rapid iteration, reference generation, extension |
| **Seedance 2.0** ββ | `doubao-seedance-2-0-260128` | Standard-quality route with first/last frames and video editing | Stable output, editing, fine control |
| **Sora 2** β | `sora-2` | Current usage-billed official API forwarding route | General video generation |
| **Sora 2 Pro** ββ | `sora-2-pro` | Current higher-quality official API forwarding route | Professional output |
| **Wan 2.7 Text-to-Video** β | `wan2.7-t2v` | DashScope-compatible async text-to-video | Prompt-based video generation |
| **Wan 2.7 Image-to-Video** β | `wan2.7-i2v` | First-frame image-to-video, optional driving audio | Animate images and audio-driven scenes |
| **Wan 2.7 Reference-to-Video** | `wan2.7-r2v` | Reference-image video generation | Subject, outfit, or style references |
| **Wan 2.7 Video Edit** | `wan2.7-videoedit` | Input video plus reference-image editing | Outfit replacement and video material edits |
| **Veo 3.1 Generate Preview** ββ | `veo-3.1-generate-preview` | Current high-quality preview route | Standard and professional video |
| **Veo 3.1 Fast Generate Preview** β | `veo-3.1-fast-generate-preview` | Current fast preview route | Rapid prototyping and batch generation |
Wan 2.7 uses the `Wan` token group, pay-as-you-go billing, and the DashScope-compatible async creation path `/wan/api/v1/services/aigc/video-generation/video-synthesis`. See [Wan 2.7 Video Generation API](/en/api-capabilities/wan-video-generation) for request parameters and polling behavior.
Seedance 2.0 uses the `SeeDance2` token group, usage-based tokens, and the async task endpoint `/seedance/api/v3/contents/generations/tasks`. See [Seedance 2.0 Video Generation API](/en/api-capabilities/seedance2-video-generation) for parameters, billing, and polling.
The July 1 retirement notice applied to older Sora website-simulation/forwarding routes, not the current usage-billed `sora-2` and `sora-2-pro` official API routes. Old `sora_video2*`, character aliases, and Veo IDs such as `veo-3.1`, `veo-3.1-fast`, `veo-3.1-fl`, and `veo3*` are for migration diagnostics only. See [Sora 2 Official API Forwarding](/en/api-capabilities/sora2/official-forward) and [Veo 3.1 Official API Forwarding](/en/api-capabilities/veo/official-forward), then verify current token groups in the console.
**Image Generation Testing Tool**
Visit [yingtu.ai](https://yingtu.ai) to experience various image generation models.
Detailed documentation:
* [GPT-Image-2 API](/en/api-capabilities/gpt-image-2)
* [Image Generation Model Guide](/en/api-capabilities/image-generation-guide)
* [GPT-Image-1 Documentation](/en/api-capabilities/gpt-image-1)
### π Claude Series (Anthropic)
Current Claude models control reasoning through the official `effort` / thinking parameters. A `-thinking` suffix is not an official Anthropic model ID, so only use such a gateway compatibility alias when it is explicitly listed in the LaoZhang API console.
#### Claude 5 Latest Series π₯
| Model Name | Model ID | Context | Features | Recommended Use |
| ----------------------- | ----------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------- |
| **Claude Opus 5** βββ | `claude-opus-5` | Check console | Newly available through LaoZhang API; supports `claude_code` / `default` groups and Anthropic / OpenAI-compatible endpoints | Latest Opus route evaluation and pre-production validation for hard tasks |
| **Claude Fable 5** βββ | `claude-fable-5` | 1M | Anthropic's latest high-capability class; global API access restored | Very complex agents, research, hard professional tasks |
| **Claude Sonnet 5** βββ | `claude-sonnet-5` | 1M | Available through LaoZhang API with a balanced cost profile | Coding agents, tool use, scaled production |
`claude-opus-5` currently lists at `$5/1M input tokens`, `$25/1M output tokens`, and `$0.5/1M cache-read tokens`. Confirm price, account access, and actual charges on the [model catalog and pricing page](/en/models), in the console, and in call logs.
#### Claude 4.8 / 4.7 / 4.6 / Haiku Compatibility Series
| Model Name | Model ID | Context | Features | Recommended Use |
| ------------------------ | ------------------- | ------- | -------------------------------------- | ------------------------------------------------------------------ |
| **Claude Opus 4.8** ββ | `claude-opus-4-8` | 1M | Previous mature Opus route | Existing production workloads and compatibility-focused hard tasks |
| **Claude Opus 4.7** β | `claude-opus-4-7` | 1M | Mature high-capability model | Complex reasoning, coding agents, long workflows |
| **Claude Sonnet 4.6** ββ | `claude-sonnet-4-6` | 1M | Balanced speed, cost, and intelligence | Code generation, analysis, long text |
| **Claude Haiku 4.5** | `claude-haiku-4-5` | 200K | Lightweight fast | Quick response |
#### Claude 4.5 Series (Classic High Performance)
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------------- | ------------------- | ------- | -------------------------------- | -------------------------------- |
| **Claude Opus 4.5** | `claude-opus-4-5` | 200K | Classic high-performance version | High-quality analysis and coding |
| **Claude Sonnet 4.5** | `claude-sonnet-4-5` | 200K | Stable coding version | Daily development |
#### Claude 4 Series (May 2025)
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------------- | ----------------- | ------- | ---------------- | ----------------- |
| **Claude 4 Sonnet** β | `claude-sonnet-4` | 200K | Stable version | Code generation |
| **Claude 4.1 Opus** | `claude-opus-4-1` | 200K | Enhanced version | High-demand tasks |
#### Claude 3.7 Series (February 2025)
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------------- | -------------------------- | ------- | -------------------- | ----------------- |
| **Claude 3.7 Sonnet** | `claude-3-7-sonnet-latest` | 200K | Legacy compatibility | General scenarios |
#### Claude 3.5 Series (Classic)
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------------- | -------------------------- | ------- | -------------------- | ----------------- |
| **Claude 3.5 Sonnet** | `claude-3-5-sonnet-latest` | 200K | Balanced performance | General scenarios |
| **Claude 3.5 Haiku** | `claude-3-5-haiku-latest` | 200K | Lightweight fast | Daily tasks |
### π Google Gemini Series
Gemini 3 and 2.5 models control Thinking through request parameters. Some gateways expose `-thinking` or `-nothinking` compatibility aliases, but those are not official Google model IDs; use them only when they appear in the LaoZhang API console.
#### Gemini 3.6 / 3.5 Series (Latest) π₯
| Model Name | Model ID | Context | Features | Recommended Use |
| --------------------------------------- | ------------------------------------ | ------- | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------- |
| **Gemini 3.6 Flash** βββ | `gemini-3.6-flash` | 1.05M | Latest stable Flash with better token efficiency, coding and agentic planning, and less verbosity | Fast multimodal work, coding, agents, knowledge work |
| **Gemini 3.5 Flash-Lite** ββ | `gemini-3.5-flash-lite` | 1.05M | Latest stable lightweight tier with low latency and low cost | Subagents, high-volume automation, classification, extraction |
| **Gemini 3.5 Flash** ββ | `gemini-3.5-flash` | 1.05M | Previous stable Flash workhorse | Existing multimodal, coding, and agent workflows |
| **Gemini 3.1 Pro Preview** ββ | `gemini-3.1-pro-preview` | 1.05M | Latest Pro preview with strong tool and agent capabilities | Advanced tasks, long text, coding agents |
| **Gemini 3.1 Pro Preview Custom Tools** | `gemini-3.1-pro-preview-customtools` | 1.05M | Optimized for custom tools and bash workflows | Agent tool use |
| **Gemini 3.1 Flash-Lite** | `gemini-3.1-flash-lite` | 1.05M | Earlier stable lightweight model | Existing lower-cost workflows |
For new projects, evaluate `gemini-3.6-flash` as the main Flash route and `gemini-3.5-flash-lite` for cost- and latency-sensitive high-volume work. Keep `gemini-3.5-flash` and `gemini-3.1-flash-lite` for existing integrations, and use only model IDs available in the console.
#### Gemini 2.5 Series (2025)
| Model Name | Model ID | Context | Features | Recommended Use |
| ------------------------- | ----------------------- | ------- | --------------------------------------------- | -------------------------- |
| **Gemini 2.5 Pro** β | `gemini-2.5-pro` | 1.05M | Stable model with strong coding and reasoning | Production workloads |
| **Gemini 2.5 Flash** β | `gemini-2.5-flash` | 1.05M | Balanced speed and cost | Quick response |
| **Gemini 2.5 Flash-Lite** | `gemini-2.5-flash-lite` | 1.05M | Lightweight, low-cost stable model | Very high-volume workloads |
#### Latest Gemini Image Models (Nano Banana) π₯
| Model Name | Model ID | Features | Recommended Use |
| ------------------------------------------------------- | ----------------------------- | ------------------------------------------------------------- | --------------------------------------------------- |
| **Gemini 3.1 Flash Image / Nano Banana 2** βββ | `gemini-3.1-flash-image` | 0.5Kβ4K, balanced quality and speed, generation and editing | High-throughput production images, advanced editing |
| **Gemini 3.1 Flash Lite Image / Nano Banana 2 Lite** ββ | `gemini-3.1-flash-lite-image` | Fixed 1K, 14 aspect ratios, low latency and low cost | Drafts, bulk assets, high concurrency |
| **Gemini 3 Pro Image / Nano Banana Pro** βββ | `gemini-3-pro-image` | Professional 4K, complex instructions, strong text and layout | Professional assets, posters, advanced editing |
| **Gemini 2.5 Flash Image / Nano Banana** | `gemini-2.5-flash-image` | Stable 1K compatibility route | Existing integrations, lower-cost generation |
For new image projects, start with `gemini-3.1-flash-image`; use `gemini-3.1-flash-lite-image` for high-volume 1K work and `gemini-3-pro-image` for complex professional output. See [Nano Banana 2](/en/api-capabilities/nano-banana2-image), [Nano Banana 2 Lite](/en/api-capabilities/nano-banana-2-lite-api), and [Nano Banana Pro](/en/api-capabilities/nano-banana-pro-image).
#### Gemini 2.0 Series (Shut Down)
`gemini-2.0-flash-001` and `gemini-2.0-flash-lite-001` were shut down in June 2026. Do not use these IDs for new integrations.
#### Gemini 1.5 Series (Historical Compatibility)
Gemini 1.5 is no longer recommended for new integrations. Prefer `gemini-3.6-flash`, `gemini-3.5-flash-lite`, `gemini-3.1-pro-preview`, or the stable `gemini-2.5-pro`.
### π xAI Grok Series
| Model Name | Model ID | Features | Recommended Use |
| --------------------------- | ------------------------------ | -------------------------------------------------------- | ------------------------------------------- |
| **Grok 4.5** βββ | `grok-4.5` | July 2026 flagship, 500K context, configurable reasoning | Coding, agents, engineering, knowledge work |
| **Grok 4.3** ββ | `grok-4.3` | 1M context, general-purpose workhorse | Long context, chat, and tool workflows |
| **Grok 4.20 Reasoning** | `grok-4.20-0309-reasoning` | Fixed 1M-context reasoning snapshot | Reasoning and version-pinned production |
| **Grok 4.20 Non-Reasoning** | `grok-4.20-0309-non-reasoning` | Direct non-reasoning snapshot | Fast generation and structured tasks |
| **Grok 4.20 Multi-Agent** | `grok-4.20-multi-agent-0309` | Multi-agent snapshot | Parallel research and role-based workflows |
Legacy IDs including `grok-4-0709`, `grok-4-fast-*`, and `grok-3` were retired by xAI and now redirect to Grok 4.3. New integrations should use `grok-4.5` or the current Grok 4.3 / 4.20 models shown in the console.
#### Grok Imagine Image and Video
| Model Name | Model ID | Features | Recommended Use |
| -------------------------------- | ---------------------------- | ---------------------------------------------------------------------------- | ------------------------------------------------ |
| **Grok Imagine Image Quality** β | `grok-imagine-image-quality` | High-quality image generation and editing | Finished assets, marketing, detailed edits |
| **Grok Imagine Image** | `grok-imagine-image` | Standard image-generation route | Fast generation and everyday assets |
| **Grok Imagine Video 1.5** β | `grok-imagine-video-1.5` | Current-generation xAI video model; text-to-video is not currently supported | Image-to-video, reference, and editing workflows |
| **Grok Imagine Video** | `grok-imagine-video` | Compatibility route that supports text-to-video | Text-to-video and existing video workflows |
These are official xAI model IDs. Confirm whether the corresponding image or video token group is available in the LaoZhang API console.
### π DeepSeek Series
| Model Name | Model ID | Context | Features | Recommended Use |
| ------------------------ | ------------------- | ------- | ------------------------------------------------------- | ---------------------------------------------- |
| **DeepSeek V4 Pro** βββ | `deepseek-v4-pro` | 1M | Latest high-capability model with Thinking and tool use | Complex reasoning, coding agents, long context |
| **DeepSeek V4 Flash** ββ | `deepseek-v4-flash` | 1M | Faster and more economical V4 route | Daily development, high volume, general tasks |
| **DeepSeek V3.2** | Check console | 128K | Previous stable hybrid-reasoning model | Existing V3.2 workloads |
| **DeepSeek R1 (05-28)** | `deepseek-r1-0528` | 128K | Legacy standalone reasoning model | Existing math and reasoning workloads |
DeepSeek will retire the `deepseek-chat` and `deepseek-reasoner` aliases on July 24, 2026. New integrations should use `deepseek-v4-pro` or `deepseek-v4-flash`; check the LaoZhang API console for current route availability.
### π Chinese Models
#### Alibaba Qwen 3.8 / 3.7 Series (Latest) π₯
| Model Name | Model ID | Context | Features | Recommended Use |
| ---------------------------- | --------------------- | ---------------------- | ---------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| **Qwen 3.8 Max Preview** βββ | `qwen3.8-max-preview` | See official plan page | Highest-generation preview; Token Plan only | Frontier evaluation; confirm plan and console access before production |
| **Qwen 3.7 Max** βββ | `qwen3.7-max` | 1M | Latest flagship with Thinking, tools, built-in search, and structured output | Complex reasoning, coding, long-running agents |
| **Qwen 3.7 Plus** ββ | `qwen3.7-plus` | 1M | Native vision-language and interactive agent capabilities | Multimodal agents, production workflows |
| **Qwen 3.6 Flash** β | `qwen3.6-flash` | 1M | Cost-effective fast model | High volume, lightweight agents, visual understanding |
Legacy aliases such as `qwq-*`, `qwen-max`, `qwen-plus`, and `qwen-turbo` may remain useful for existing projects, but they no longer represent the latest Qwen generation. Check the LaoZhang API console for callable IDs.
#### Alibaba Qwen Coder 3 Series (Coding) π₯
| Model Name | Model ID | Parameters | Features | Recommended Use |
| ----------------------- | -------------------------------- | ----------------- | ------------------ | --------------- |
| **Qwen3 Coder 480B** ββ | `qwen3-coder-480b-a35b-instruct` | 480B (35B active) | Large coding model | Complex coding |
| **Qwen3 Coder Plus** β | `qwen3-coder-plus` | - | Enhanced coding | Standard coding |
#### Other Chinese Models
| Model Name | Model ID | Context | Features |
| ---------------------------- | --------------------------- | ------- | --------------------------------------------------------------------------- |
| **Kimi K3** βββ | `kimi-k3` | 1.05M | July 2026 flagship with always-on Thinking, vision, and long-horizon coding |
| **Kimi K2.7 Code** ββ | `kimi-k2.7-code` | 256K | Thinking-only model for long-horizon software engineering |
| **Kimi K2.7 Code Highspeed** | `kimi-k2.7-code-highspeed` | 256K | Higher-speed K2.7 Code route |
| **Llama 4 Maverick** | `llama-4-maverick` | - | Established open-source compatibility route |
| **Seedream 4.5** ββ | `seedream-4-5-251128` | - | LaoZhang API's currently documented high-quality image route |
| **Doubao 1.5 Vision Pro** β | `Doubao-1.5-vision-pro-32k` | 32K | Multimodal |
| **Gemma 3 12B** | `gemma-3-12b` | - | Google open source |
| **GLM-5.2** ββ | `glm-5.2` | 1M | Provider-native 1M context and 128K max output; gateway limits may be lower |
| **MiniMax M3** β | `MiniMax-M3` | 1M | Provider-native multimodal 1M context; some gateways expose only 192K |
Moonshot's official model IDs do not include a `kimi/` channel prefix; use such a prefix only when a gateway explicitly lists it. The Kimi K2 series has been discontinued and is no longer recommended for new integrations.
## π° Pricing Information
### Billing Method
* **Pay-as-you-go**: Charged based on actual token usage
* **No minimum charge**: Use what you pay for, balance never expires
* **Real-time deduction**: Fees deducted immediately after each call
### Pricing Advantage
* Direct from official sources with competitive rates
* Enterprise pricing terms available through support
* New users receive \$0.5 free trial credit
### View Real-time Pricing
Visit [LaoZhang API Console Pricing Page](https://api2.laozhang.ai/account/pricing) to view the latest prices for all models.
## π οΈ Usage Recommendations
### Model Selection Guide
**Programming Development**
* Primary: GPT-5.6 Sol, Claude Opus 5, Claude Sonnet 5, Grok 4.5
* Alternatives: GPT-5.6 Terra, Claude Opus 4.8, Gemini 3.6 Flash, Gemini 3.1 Pro Preview, Qwen3 Coder 480B
**Content Creation**
* Primary: GPT-5.6 Sol, Claude Opus 5, Claude Sonnet 5
* Alternatives: GPT-5.6 Terra, Grok 4.5, Gemini 3.6 Flash, Qwen 3.7 Max, Kimi K3
**Quick Response**
* Primary: Gemini 3.6 Flash, Claude Sonnet 5, Claude Haiku 4.5
* Alternatives: GPT-5.6 Terra, Grok 4.3, Gemini 3.5 Flash, Gemini 2.5 Flash
* Cost-focused: Gemini 3.5 Flash-Lite, GPT-5.6 Luna, Gemini 2.5 Flash-Lite
**Image Generation**
* Top quality: GPT-Image-2, Gemini 3 Pro Image (Nano Banana Pro), Flux Kontext Max
* Balanced: Gemini 3.1 Flash Image (Nano Banana 2), GPT-Image-2 VIP
* Cost-focused: Gemini 3.1 Flash Lite Image (Nano Banana 2 Lite), Gemini 2.5 Flash Image
**Video Generation**
* Primary: Seedance 2.0, `veo-3.1-generate-preview`, Wan 2.7, Sora 2 Pro
* Fast iteration: Seedance 2.0 Fast, `veo-3.1-fast-generate-preview`
* OpenAI route: current official forwarding uses `sora-2` / `sora-2-pro`; old `sora_video2*` and character aliases are only for migration diagnostics
**Long Text Processing**
* Primary: Claude Opus 5, Claude Fable 5, GPT-5.6 Sol, Gemini 3.1 Pro Preview
* Alternatives: GPT-5.6 Terra, Claude Sonnet 5, Grok 4.3, Gemini 3.6 Flash
### Cost Optimization Tips
1. **Tiered Usage**: Use lower-cost models for simple tasks, premium models for complex ones
2. **Test and Optimize**: Test with smaller models first, then scale up as needed
3. **Batch Processing**: Choose Nano or Mini versions for large volumes of similar tasks
4. **Cache and Reuse**: Cache results for repeated queries
## π Related Resources
* [Model Comparison Testing](https://yingtu.ai) - Image generation comparison
* [Real-time Pricing](https://api2.laozhang.ai/account/pricing) - Latest pricing information
* [Model Catalog and Pricing](/en/models) - Current `claude-opus-5` price, cache-read price, groups, and compatible endpoints
* [API Documentation](/en/api-manual) - Detailed interface specifications
* [Quick Start](/en/getting-started) - Integration guide
* [OpenAI GPT-5.6 Model Directory](https://developers.openai.com/api/docs/models) - Official Sol, Terra, and Luna model IDs and context details
* [OpenAI GPT-5.6 Guide](https://developers.openai.com/api/docs/guides/latest-model) - Alias, reasoning effort, and Pro-mode details
* [OpenAI Image Generation Guide](https://developers.openai.com/api/docs/guides/image-generation) - GPT-Image-2 size, quality, and output constraints
* [Anthropic Claude Fable 5 Announcement](https://www.anthropic.com/news/claude-fable-5-mythos-5) - Official model ID and availability scope
* [Anthropic Claude Opus 4.8 Announcement](https://www.anthropic.com/news/claude-opus-4-8) - Official model ID and 1M context details
* [Anthropic Claude Sonnet 5 Announcement](https://www.anthropic.com/news/claude-sonnet-5) - Official model ID and release status
* [Google Gemini 3.6 Flash](https://ai.google.dev/gemini-api/docs/models/gemini-3.6-flash) - Official capabilities and context details for the latest stable Flash model
* [Google Gemini 3.5 Flash-Lite](https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash-lite) - Official details for the latest stable lightweight model
* [Google Gemini 3.1 Flash Lite Image](https://ai.google.dev/gemini-api/docs/models/gemini-3.1-flash-lite-image) - Official 1K and 14-aspect-ratio details for Nano Banana 2 Lite
* [xAI Grok 4.5 Model Page](https://docs.x.ai/developers/models/grok-4.5) - Official model ID and context details
* [xAI Grok Imagine Image Quality](https://docs.x.ai/developers/models/grok-imagine-image-quality) - Current high-quality image model and aliases
* [xAI Grok Imagine Video 1.5](https://docs.x.ai/developers/models/grok-imagine-video-1.5) - Current video model and its image-input-only limitation
* [DeepSeek API Changelog](https://api-docs.deepseek.com/updates/) - V4 models and legacy alias retirement dates
* [Alibaba Cloud Model Studio Text Models](https://help.aliyun.com/en/model-studio/text-generation-model/) - Qwen, DeepSeek, Kimi, and GLM parameters
* [Moonshot Kimi Model List](https://platform.kimi.ai/docs/models) - Official K3 and K2.7 Code IDs and context limits
* [Zhipu GLM-5.2 Documentation](https://docs.bigmodel.cn/cn/guide/models/text/glm-5.2) - Native context and output limits
* [MiniMax Model Release Notes](https://platform.minimaxi.com/docs/release-notes/models) - MiniMax M3 release and capabilities
* [Sora 2 Official API Forwarding](/en/api-capabilities/sora2/official-forward) - Current LaoZhang API route and token requirements
* [Veo 3.1 Official API Forwarding](/en/api-capabilities/veo/official-forward) - Current LaoZhang API model IDs and token requirements
Model list is continuously updated. We promptly add newly released excellent models. For specific model needs or bulk requirements, please contact support.