OpenRouter 模型 ↗ · 2026-11-04
tencent/hy4
模型发布 OpenRouter 模型 原文语言:中文
OpenRouter 于 2026-11-04 收录该模型(收录时间不等于模型正式发布日期);上下文:1,048,576 tokens;输入价:$0.4170/百万 tokens;输出价:$1.2505/百万 tokens;定位(来源描述):The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview,…
全文
Tencent: Hy4
tencent/hy4
Model weights
Compare
The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity tasks. It also introduces visual understanding for the first time.
Modalities
In / out price
50% off
$0.417 / $1.251per 1M
Context
1.0M
Released
Nov 4, 2026
Compare
Providers
This model is hosted by one provider. OpenRouter forwards every request without routing.
Latency / throughput
50% off
$0.834$0.417$2.501$1.251$0.042$0.0212.82s66 tok/s
100.00%
Provider50% offInput / Output /M
Input $0.417 / Output $1.25150% off
66 tok/s100.00% uptime
Pricing
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Weighted average
Input
$0.08431/M tokens
Output
$2.472/M tokens
Hy4 — Price history
Price history
No effective price data for this selection
$0.0843$2.4794.6%100.0%
ProviderInput / Output /M
Token share
Effective in $0.0843 / Effective out $2.47
Token share
100.0%
Performance
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
Throughput
66tok/s
P50, best across providers
Latency
2.82s
P50, best provider
Reasoning effort
AutoExacto benchmarks
GPQA DiamondTencent Cloud83.2%
Uptime
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.88%
Availability over the last 3 days
Last 72 hours
Availability 99.88%
3 Days Ago2 Days AgoYesterdayNow
Availability over the last 24 hours
OpenRouter Availability
99.83%
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.
Benchmarks
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
Benchmark score summary for Tencent: Hy4 (Artificial Analysis, Design Arena and Vals AI)
SourceBenchmarkScore
Artificial AnalysisTencent Hy4 Intelligence Index34.6
Artificial AnalysisTencent Hy4 AA-Omniscience Accuracy50.0%
Artificial AnalysisTencent Hy4 AA-Omniscience Non-Hallucination Rate10.9%
Design ArenaHy4 Models Arena 3D Elo1251
Design ArenaHy4 Models Arena Code Categories Elo1253
Design ArenaHy4 Models Arena Game Development Elo1271
Design ArenaHy4 Models Arena UI Component Elo1194
Design ArenaHy4 Models Arena Website Elo1250
Vals AIHy4 EMB v1.166.2%
Vals AIHy4 Finance Agent v2.164.7%
Vals AIHy4 Vals Finance Index58.1%
Vals AIHy4 HLAB-Vals-1.112.5%
Vals AIHy4 Vals Legal Index70.6%
Vals AIHy4 Legal Research Bench41.3%
Vals AIHy4 Tax Agent Bench34.4%
Vals AIHy4 Terminal-Bench 4.039.4%
Vals AIHy4 Vals Index59.4%
Vals AIHy4 Vibe Code Bench v1.181.0%
Apps
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
1.
Claude Code
Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.
258Btokens
2.
Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
44.5Btokens
3.
Strix
Open-source AI hackers that find and validate vulnerabilities in your code, APIs, and infrastructure, and open fix PRs.
14.3Btokens
4.
DeepSeek Harness
new
13.3Btokens
5.
Nous Research API
Nous Research is an AI research team focused on advancing open-source AI.
9.83Btokens
Rest of OpenRouter
1.85T
Top 5 Apps
9.72B
Activity
Token volume and request traffic to this model over time.
API
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
Not currently available through EU in-region routing. In-region routing docs
Frequently asked questions
The official release of Hy4 delivers comprehensive and substantial improvements over Hy4 Preview, with particularly strong gains in agentic capabilities and performance across a wide range of long-horizon, real-world productivity tasks. It also introduces visual understanding for the first time.
Hy4 costs $0.417/M input tokens and $1.251/M output tokens, with separate rates for Cache Read at $0.021/M tokens.
Hy4 has a 1,048,576 token context window. It supports up to 65,536 completion tokens.
Yes. Hy4 accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
Hy4 accepts text and images as input and returns text.
Hy4 is served by 1 provider on OpenRouter: Tencent Cloud.
Hy4 preview, Hy-MT2-1.8B, Hy-MT2-30B-A3B and 3 more are other text models from Tencent.
Hy4 was released on November 4, 2026.
正文以来源语言显示。