none reviewed yet · 1 benchmark observation · latest capture Sep 8, 2026
Inspect the evidenceTencent · MODEL PROFILE
Tencent Hy3
High-volume, cost-capped code and content generation where design quality is not critical
Saved editorial fit for your current task lens; not a benchmark result
How scoring worksVerified rates (Sep 8, 2026) · 100 runs / month at your token settings
Adjust workloadAt a glance
9 facts- Provider
- Tencent
- Type
- API model
- Access
- Tencent Cloud, WorkBuddy, CodeBuddy, self-hosting
- Context
- 262,144 tokens
- Maximum output
- n.a. on fetched pages
- License / weights
- Open weights
- Input modalities
- text
- Output modalities
- text
- Pricing notes
- $0.14 in / $0.55 out per 1M (Artificial Analysis) - the cheapest model in this set
The practical trade-offs
Documented strengths
- Cheapest tokens in this set ($0.14/$0.55) with Apache 2.0 weights
- 94.8 tok/s with 2.76 s TTFT - genuinely responsive
- SWE-bench Verified 74.4% is respectable for the price tier
- Free in Tencent's own dev products until Sep 30, 2026
Limitations to check
- Design Arena Website 1197 (rank 64) is the weakest visual score in this set
- 256k context is the smallest here
- AA Intelligence Index 33 is low
- The only SWE-bench row is labelled 'Preview', so it may not match GA hy3
Follow the source
5 source linksEvidence & conditions
5Source links
NoneReviewed records
1Benchmark runs
Sep 8, 2026Latest capture
- Evidence source 1
artificialanalysis.aiDate not recorded - Evidence source 2
arena.aiDate not recorded - Evidence source 3
benchlm.aiDate not recorded - Evidence source 4
benchlm.aiDate not recorded - Evidence source 5
tencent.comDate not recorded
Benchmark observations
Arena WebDev Frontend Inspect conditions
- name
- Arena WebDev Frontend
- value
- 1517
- unit
- arena_score
- source Url
- https://arena.ai/leaderboard/code/webdev/frontend
- checked At
- 2026-09-08T19:28:18Z
- attribution
- benchmark-publisher
- configuration
- Exact row hy3; source results dated 2026-09-05; human preference; confidence interval +/-13; 2576 votes. This is a checked source snapshot, not a newly run September 8 evaluation.
Source reports, measured results, and editorial assessments are separate evidence types. Source dates do not imply a new benchmark run. Capture dates are shown when recorded; otherwise we say the capture date was not recorded. Recorded dates are compared with this catalogue snapshot (Sep 11, 2026), not today’s date.
Method AIS-V2
How this page is assessed
Quality claims follow Method AIS-V2.
Money never buys rank — commercial relationships cannot change rank, eligibility, evidence quality, or publication timing.
Submit or claim a listing is free and does not skip the editorial queue.
Private · this device only