Hy4 flash
by Tencent · tencent/hy4-flash · #26 cheapest of 457 paid models
Prices updated Nov 18, 2026, 12:28 AM UTC · refreshed nightly
Hy4 flash is a cost-efficient variant of [Hy4](https://openrouter.ai/tencent/hy4) from Tencent, served with roughly a quarter of the flagship's parameter count for lower latency and cost. It targets code assistance, agent...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.035
per 1M tokens
Output
$0.14
per 1M tokens
Blended (3:1)
$0.061
3 input : 1 output
Cache read
$0.0015
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Hy4 flash with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.061 | $0.003711 | — |
| 50% | $0.049 | $0.002035 | −45% |
| 90% | $0.039 | $0.000696 | −81% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $5.96e-5 | $0.0597 |
| Document summary | 8,000 | 1,000 | $0.000421 | $0.4205 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.001331 | $1.33 |
| Long-context analysis | 150,000 | 5,000 | $0.005953 | $5.95 |
Performance
Intelligence Index
—
AA composite quality
Coding Index
—
Math Index
—
Agentic Index
—
Output speed
—
median
Time to first token
—
median
Time to first answer
—
after reasoning tokens
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,234 |
| 3D | 1,215 |
| Game dev | 1,225 |
Specs
Context window
1.05M
tokens
Max output
66K
tokens
Input modalities
text, image
Tokenizer
Other