Balanced — strong quality at a mid priceOpen Weights

Hy4 preview Pricing

Tencent · tencent-hy4-preview

ShareX Facebook LinkedIn

Input / 1M tokens

$0.834

Output / 1M tokens

$2.50

Context window

1,048,576 tokens

≈ 1,398 pages of text

Max output

64k tokens

What it costs in practice

Typical request (1,200 in + 400 out tokens)$0.002

1,000 requests / month$2.00/mo 10,000 requests / month$20.01/mo 100,000 requests / month$200.12/mo

Estimate your own workloadOpens the calculator with Hy4 preview preloaded — adjust volume and token counts there.

Price history

List price per 1M tokens since we started tracking in Aug 2026. Each step marks a price change — hover a point for the exact date and rate.

What it excels at

Supports 1M-token context and 64k output, enabling extended document and multi-turn analysis workloads. Open weights allow self-hosted deployment at listed per-token rates.

The business tradeoff

Preview status implies limited production validation and potential gaps in instruction following or edge-case stability. Output-token pricing and hosting variance remain operator-dependent.

Cheaper from Tencent

Hy-MT2-30B-A3BTencent$0.0002/ typical request Hy-MT2-1.8BTencent$0.0001/ typical request

Same tier from other providers

GPT-5.4 MiniOpenAI$0.0027/ typical request Claude Sonnet 4.6Anthropic$0.0096/ typical request Llama 4 MaverickMeta$0.0005/ typical request

Head-to-head

Hy4 preview vs GPT-5.4 Mini →Hy4 preview vs Claude Sonnet 4.6 →Hy4 preview vs Llama 4 Maverick →

Vendor list rates, as of Oct 26, 2026 · source: openrouter · per-request examples assume 1,200 input + 400 output tokens.