LLM Pricing
ModelsLeaderboardsToolsProvidersReleasesArticles

© 2026 LLM Pricing

About
·Contact
·Privacy
·RSS
← Model list
hu

openPangu 2.0 Pro

huawei·huawei/openpangu-2.0-pro·GA·Open weights·pangu series
Huawei's 505B MoE flagship (18B active) for long-horizon agents and 512K context work, Ascend-native with slow/fast thinking

Specs & pricing

Input / output per 1M tokens
Official price
—
Lowest paid·Huawei Cloud MAASGateway
$0.48 / $2.17
Blended $0.90
Context
524,288
Output limit
131,072
Knowledge cutoff
2026-06
Released / updated
2026-07-31 / 2026-07-31
Capabilities
✓ Reasoning✓ Tool use? Structured output✓ TemperatureAttachments
Modalities
Text
Weights
Hugging Face

openPangu 2.0 Pro by huawei is currently listed from a single provider. No official API price is listed for it; the prices below come from third-party channels.

The context window is 524,288 tokens, with an output limit of 131,072 tokens. It supports reasoning and tool use. The weights are open, so it can also be self-hosted or served through a gateway of your choice. Its training knowledge cuts off at 2026-06.

Available at 1 providers1 with public prices

  • ProviderInput / Output
  • Huawei Cloud MAASGateway
    $0.48 / $2.17
    tiered >32K: $0.72
    Cache read $0.12Context 524,288Output limit 131,072
ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Huawei Cloud MAASGateway$0.48
tiered >32K: $0.72
$2.17$0.12—524,288131,072

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Reasoning control

Toggle (on / off)effort = loweffort = higheffort = max

Interleaved thinking (reasoning between tool calls) is declared by 1 of 1 providers.

Your usage cost

Common scenarios
1Huawei Cloud MAAS$161.47
The cheapest paid channel is the only channel.

Price historyone sample accumulated per data sync

Input list —Output list —Min blended $0.90

Price history accumulates from each data sync; currently only 1 sample(s) (2026-11-15). Each future sync adds a point, and once accumulated a line is drawn here.

Benchmark4 items

NameConditionsScoreMetricSource
AIME 2026variant: thinking95.4Avg@16Source ↗
SWE-bench Verifiedvariant: thinking68.5Avg@3Source ↗
LiveCodeBench V6variant: thinking85.7Avg@3Source ↗
GPQA-Diamondvariant: thinking87.9Avg@4Source ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Related models

huopenPangu 2.0 Flashsame series—
Data partly from models.dev (MIT)