model release

Huawei Cloud opens openPangu-2.1-Pro on MaaS: 512K context, output at 3.2 yuan per 1M tokens

TL;DR

Huawei Cloud's release notes list openPangu-2.1-Pro (version 20261115) on its MaaS platform with a 512K-token context window. Pricing is 0.8 yuan per million input tokens and 3.2 yuan per million output tokens. The preset service is available only in CN Southwest-Guiyang1.

2 min read
0

Huawei Cloud has opened openPangu-2.1-Pro on its MaaS platform with a 512K-token context window and output priced at 3.2 yuan per million tokens. The release notes list the version as 20261115.

Specs and features

  • Context window: 512K tokens
  • Supported features: function calling, prefix completion, prefix caching and chain-of-thought
  • Availability: the preset service is offered only in the CN Southwest-Guiyang1 region

Pricing

Per million tokens, according to Huawei Cloud's price table:

Token type Price (yuan)
Input 0.8
Output 3.2
Cache-hit input 0.016

The table shows no peak or off-peak rates and no tiers by request length.

The previous openPangu-2.0-Pro charged 14.5 yuan per million output tokens for requests under 32K tokens and 17.6 yuan above that. IT Home reports that the new output price is about 18% to 22% of the old one.

Platform plans and retirements

Huawei Cloud also plans to bring the model into its AgentArts agent platform, CodeArts coding tools and OfficeAce office product.

The launch follows a quick turnover on the platform. Huawei Cloud has scheduled openPangu-2.0-Pro and openPangu-2.0-Flash to go offline on November 30. Its retirement notice pointed users to Zhipu's GLM-5.3 and DeepSeek-V4.1-Flash-1017 as replacements.

Sources: IT Home, Huawei Cloud openPangu-2.1-Pro Quick Start, Huawei Cloud MaaS Text Model Pricing.

What this means

The new model uses a single price for each token type, so cost estimates no longer depend on whether a request is under or over 32K tokens, as they did for openPangu-2.0-Pro. The 0.016 yuan cache-hit input rate is paired with prefix caching, which makes repeated long prefixes the cheapest way to use the 512K window. Teams running openPangu-2.0-Pro or openPangu-2.0-Flash have until November 30 to migrate. Availability only in CN Southwest-Guiyang1 is the practical constraint for now.

Related Articles

model release

Huawei Cloud to unveil CloudRobo-WM1.2, its first Ascend-native world model, on November 16

Huawei Cloud will launch CloudRobo-WM1.2, an embodied world model for robots, at an online event on November 16 (14:00 to 15:30 Beijing time). Its poster bills the model as the first Ascend-native world model from the Industry AI Dream Factory. Parameter counts, benchmarks, availability and pricing have not been disclosed.

model release

Google launches Gemini 4 Hydrogen: 1M-token context, $1.50 input and $7.50 output per 1M tokens

Google announced Gemini 4 Hydrogen on November 12, 2026, with native 1 million-token input and up to 1 million output tokens. It costs $1.50 per million input tokens and $7.50 per million output tokens, and Google claims it scores 51 on the Artificial Analysis Intelligence Index at its default high thinking level.

model release

Qwen4 Max launches with 1M-token context and $1.50/$4.50 per 1M token pricing

Qwen4 Max, the flagship model in Alibaba's Qwen4 series, is listed on OpenRouter with a 1M-token context window. It accepts text, image and video input, returns text, and is priced at $1.50 per 1M input tokens and $4.50 per 1M output tokens.

model release

Bland Releases Tofu-1, a Full-Duplex Speech-to-Speech Model for Noisy Phone Calls

Bland has released Tofu-1, a full-duplex speech-to-speech model built for noisy, high-stakes phone calls. It is in enterprise research preview, with no published benchmarks, size or pricing yet, according to MarkTechPost.

Comments

Loading...