Tencent: Hy3
tencent/hy3Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use.
Modalities
In / out per 1M$0.13 / $0.53
Context256K
ReleasedJul 6, 2026
Providers
5 providers serve this model upstream.
| Provider | Input /M | Output /M | Cache read /M | Latency P50 | Throughput P50 | Uptime |
|---|---|---|---|---|---|---|
| $0.13 | $0.53 | $0.03 | 2.79s | 50 tps | 100% | |
| $0.14 | $0.58 | $0.04 | 2.22s | 59 tps | 100% | |
| $0.14 | $0.58 | $0.04 | 2.48s | 42 tps | 100% | |
| $0.15 | $0.64 | $0.04 | 2.38s | 53 tps | 99.91% | |
| $0.20 | $0.80 | $0.05 | 2.18s | 71 tps | 99.48% |
Market-wide measurements, refreshed every 15 minutes. They describe who serves Hy3 in general, not how ProjectCOZY routes your requests.
Pricing
List rates per 1M tokens — what Hy3 costs pay-as-you-go elsewhere, not what you pay here.
Input
$0.13
per 1M tokens
Output
$0.53
per 1M tokens
Cache read
$0.03
per 1M cached tokens
Included in every pass
Hy3 is included in every ProjectCOZY pass at no extra cost — one flat rate, no per-token metering. The rates above are there for comparison.
Performance
Time to first token and output speed across every provider, over the last 3 days.
0.0s5.0s10.0s15.0s20.0s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
GMICloud Tencent Cloud Phala NovitaAI AtlasCloud
0 tok/s200 tok/s400 tok/s600 tok/s800 tok/s
Oct 11, 12 AMOct 11, 8 AMOct 11, 4 PMOct 12, 12 AMOct 12, 8 AMOct 12, 4 PMOct 13, 12 AMOct 13, 8 AMOct 13, 4 PMOct 14, 12 AMOct 14, 8 AM
GMICloud Tencent Cloud Phala NovitaAI AtlasCloud
Uptime
100% of requests were served successfully over the last 3 days, across all providers.
Last 3 daysAvailability 99.84%
Oct 11, 2:00 PMNow
Availability over the last 24 hoursAvailability 99.85%
100% 98% 97% 95%
Oct 13, 2:40 PMNow
Apps
Tools that most often run Hy3, by tokens this week.
Activity
13324.9B tokens processed market-wide in the last month.
0200.0B400.0B600.0B800.0B
Sep 14Sep 19Sep 24Sep 29Oct 4Oct 9Oct 14
Tokens
FAQ
Which model id do I use for Hy3?
tencent/hy3. Send it as the model field from any OpenAI-compatible client, against http://192.168.1.55/v1.
How much context does Hy3 hold?
262,144 tokens per request, of which up to 128,000 can be the response.
What can it accept and return?
Input: text. Output: text.
What does Hy3 cost on a pass?
Nothing beyond the pass itself. Pay-as-you-go elsewhere it lists at $0.13 per 1M input tokens and $0.53 per 1M output tokens.
Who built Hy3?
Tencent, in the Hy3 family.
Explore
Other models your pass can call.
