MAI Code 2 vs openPangu 2.0 Pro vs Hy4
No independent capability benchmark covers every model here yet. The score uses price and context window.
Price in rating: Included · Exclude price
Microsoft
MAI Code 2
53/100- ECI—
- Price$0.50 / $2.50
- Context1.05M
Huawei
openPangu 2.0 Pro
51/100- ECI—
- Price$0.48 / $2.17
- Context524K
Tencent
Hy4
49/100- ECI—
- Price$0.834 / $2.50
- Context1.05M
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (MAI Code 2 53/100, openPangu 2.0 Pro 51/100, Hy4 49/100) compare the available specifications and price. The score weighs price 71%, context window 29%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceopenPangu 2.0 ProopenPangu 2.0 Pro $0.903 · MAI Code 2 $1.00 · Hy4 $1.25 per 1M tokens (3:1 blend)
- Longest contextMAI Code 2 and Hy4MAI Code 2 1,050,000 · Hy4 1,048,576 · openPangu 2.0 Pro 524,288 tokens
- Widest inputsMAI Code 2 and Hy4MAI Code 2: Text, Images · openPangu 2.0 Pro: Text · Hy4: Text, Images
- Self-hostingopenPangu 2.0 ProPublishes downloadable weights
| Measure | Weight | MAI Code 2 | openPangu 2.0 Pro | Hy4 |
|---|---|---|---|---|
| Price | 71% | 48 | 50 | 43 |
| Context window | 29% | 65 | 54 | 65 |
| Specs & price only | 100% | 53/100 | 51/100 | 49/100 |
Left out because at least one model lacks the data: capability and inputs & features. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | openPangu 2.0 ProHuawei | ||
|---|---|---|---|
| Task | Coding | — | — |
| Published capabilities | Mixture-of-experts coding model for long-running engineering tasks and multi-file changes. | Huawei's 505B MoE flagship (18B active) for long-horizon agents and 512K context work, Ascend-native with slow/fast thinking | A next-generation multimodal productivity model with significantly enhanced Agent and complex task execution capabilities. |
| Model size | 80B active parameters | — | — |
| Access | GitHub Copilot | — | — |
| Billing terms | Promotional GitHub Copilot token rates until May 17, 2027. | — | — |
| Sources | Publisher documentation · Additional documentation · Pricing documentation | — | — |
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.50 | $0.48 (best) | $0.834 |
| Output | $2.50 | $2.17 (best) | $2.50 |
| Cached input | $0.025 (best) | — | $0.042 |
| Blended (3:1) | $1.00 | $0.903 (best) | $1.25 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official GitHub Copilot API | Median of 1 providers | Official Tencent TokenHub API |
| Limits | |||
| Context window | 1,050,000 tokens (best) | 524,288 tokens | 1,048,576 tokens |
| Max output | 128,000 tokens | 131,072 tokens (best) | 65,536 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | Yes |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | Yes | Yeshigh |
| Tool calling | Yes | Yes | Yes |
| Structured output | — | No | No |
| Availability | |||
| Weights | Proprietary | Open | Proprietary |
| API model ID | — | — | hy4 |
| API providers | — | 1 | 12 (best) |
| Released | Nov 17, 2026 | Jul 31, 2026 | Nov 4, 2026 |
| Knowledge cutoff | — | Jun 2026 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
MAI Code 2$10.00
- openPangu 2.0 Pro$9.15
Hy4$13.34
Which should you choose?
Which is better: MAI Code 2, openPangu 2.0 Pro or Hy4?
There is not enough shared capability evidence to name an overall winner. The scores (MAI Code 2 53/100, openPangu 2.0 Pro 51/100, Hy4 49/100) compare the available specifications and price. The score weighs price 71%, context window 29%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, MAI Code 2, openPangu 2.0 Pro or Hy4?
openPangu 2.0 Pro is cheaper at $0.48 input / $2.17 output per million tokens (median across 1 API provider). MAI Code 2 costs $0.50 input / $2.50 output per million tokens (official GitHub Copilot API price); Hy4 costs $0.834 input / $2.50 output per million tokens (official Tencent TokenHub API price). At a typical mix of three input tokens to one output token, that is $0.903 per million tokens for openPangu 2.0 Pro versus $1.00 for MAI Code 2 (1.1× as much) and $1.25 for Hy4 (1.4× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. MAI Code 2 has not been scored yet, openPangu 2.0 Pro has not been scored yet and Hy4 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for MAI Code 2, openPangu 2.0 Pro and Hy4 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
MAI Code 2 and Hy4 have the largest context windows (1,050,000 and 1,048,576 tokens), against 524,288 for openPangu 2.0 Pro. Maximum output per response: MAI Code 2 up to 128,000, openPangu 2.0 Pro up to 131,072, Hy4 up to 65,536 tokens.
Which can read images, PDFs, audio or video?
MAI Code 2 accepts text and images; openPangu 2.0 Pro accepts text; Hy4 accepts text and images. MAI Code 2 handles the widest range of inputs.
Are any of these open source?
openPangu 2.0 Pro publishes its weights and can be self-hosted; MAI Code 2 and Hy4 is proprietary.
Which is newer?
MAI Code 2 is the newest, released Nov 17, 2026. Hy4 came out Nov 4, 2026; openPangu 2.0 Pro came out Jul 31, 2026. Knowledge cutoff: openPangu 2.0 Pro Jun 2026.
How do you decide the winner?
This rating uses price (71%, blended price per million tokens (3 input : 1 output), log scale); context window (29%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 19, 2026.