text model comparison
Hy4 vs Kimi K3.1 Pro
Hy4 and Kimi K3.1 Pro are both current text models, from Tencent and Moonshot Kimi. Hy4 is 58% cheaper per 1M input tokens than Kimi K3.1 Pro, $0.834 vs $2. In Tencent's Hy4 launch table, Hy4 scores 74.9% on DeepSWE v1.1 and Kimi K3.1 Pro 76.3%.
- Lower price per 1M input tokens
- Hy4$0.834 against $2
- Newer release
- Hy4Nov 2026 against Oct 2026
Data verified November 5, 2026 · Dataset 2026-11-05
Side by side
Hy4 and Kimi K3.1 Pro, line by line
Prices in effect today, compared at the same tier wherever both list one. The better value in each row is highlighted.
| Fact | Hy4 | Kimi K3.1 Pro |
|---|---|---|
| Price | ||
| Price per 1M input tokens | $0.834 (better) | $2 |
| Price per 1M output tokens | $2.501 (better) | $10 |
| Independent benchmarks | ||
| Arena WebDev | Not measured | 1695 Elo (#7 of 46) |
| Specs | ||
| Provider | Tencent | Moonshot Kimi |
| Release date | November 4, 2026 | October 23, 2026 |
| Access | API only | API and open weights |
| Context window | Not reported | 1,048,576 tokens |
| Max output | Not reported | 1,048,576 tokens |
| Input | text, image | text, image, video |
| Output | text | text |
No independent benchmark has scored both models yet. Each rank counts the current models scored on that benchmark.
Scores:Arena · 2026-11-03
Provider results
What the labs report
Results tables a lab published with one of its models, where it lists both Hy4 and Kimi K3.1 Pro. These are the lab's own figures, not independent measurements, and each table is read on its own: labs run benchmarks with different harnesses and budgets.
Tencent, Hy4 launch table, November 4, 2026Tencent · 2026-11-05
| Benchmark | Hy4 | Kimi K3.1 Pro |
|---|---|---|
| DeepSWE v1.1 | 74.9% | 76.3% (better) |
| SWE-Marathon v1.1 | 43.8% | 53.8% (better) |
| ProgramBenchalmost solved | 30.0% (better) | 24.5% |
| Terminal-Bench 4.0 | 44.4% | 66.8% (better) |
| FrontierCode 2.0 | 31.0% (better) | 25.5% |
| APEX-Agents v1.1pass@1 | 75.2% (better) | 68.2% |
| Agents' Last Exam v1.1pass rate | 41.0% | 43.4% (better) |
| AutomationBench v1.0.6 | 56.2% | 65.6% (better) |
| HLE-Fullwith tools | 59.1% | 61.8% (better) |
| Terminal-Bench-Science 0.1 | 19.8% | 41.7% (better) |
| PostTrainBench v1.2 | 35.2% | 36.2% (better) |
| OSWorld 2.1full, partial score | 78.9% | 79.6% (better) |
| MMMU-Pro | 78.6% | 83.4% (better) |
| ZeroBench mainwith tools, pass@5 | 55.4% | 56.9% (better) |
Tencent's appendix to the Hy4 announcement, every model at its highest reasoning setting. Where Tencent lists a lab's reported score beside its own run, the score Tencent measured is shown. DeepSeek-V4.1-Flash is the 1017 build. Left out here: Tencent's internal benchmarks, the Hy4 preview column and GLM-5.3 on HLE, which Tencent ran on the text-only subset.
Charts
Against every current text model
Where Hy4 and Kimi K3.1 Pro land among the current models of their task, on every independent benchmark, on price and on context.
Where each stands
Only Kimi K3.1 Pro has a current Arena WebDev score: #7 of 46. Each bar is the share of current models it beats.
Independent benchmarks
Arena WebDev
Hy4: Not scored
Kimi K3.1 Pro:#7of 46
1,695 Elo · ahead of 87%
Price and context
Price per 1M input tokens
Hy4:#21of 55
$0.834 · cheaper than 63%
Kimi K3.1 Pro:#30of 55
$2 · cheaper than 24%
Price per 1M output tokens
Hy4:#20of 55
$2.501 · cheaper than 65%
Kimi K3.1 Pro:#35of 55
$10 · cheaper than 26%
Context window
Hy4: Not listed
Kimi K3.1 Pro:#11of 64
1.05M tokens · longer than 67%
Cost
What each costs
Example workloads at each provider's listed rates, then every other rate either one lists. The cheaper figure in each row is highlighted.
The same work, priced
For 10M input and 2M output tokens, Kimi K3.1 Pro costs 3.0 times as much as Hy4, at each provider's listed rates.
10M input and 2M output tokens
Kimi K3.1 Pro costs 3.0× as much
- Hy4
- $13.34
- Kimi K3.1 Pro
- $40
The same, with 9M of the input tokens cached
Kimi K3.1 Pro costs 3.8× as much
- Hy4
- $6.21
- Kimi K3.1 Pro
- $23.80
| Workload or rate | Hy4 | Kimi K3.1 Pro |
|---|---|---|
| Example workloads | ||
| 10M input and 2M output tokens | $13.34 (better) | $40 |
| The same, with 9M of the input tokens cached | $6.21 (better) | $23.80 |
| Other listed rates | ||
| 1M input tokens, cached | $0.042 (better) | $0.20 |
| 1M input tokens, 5-minute cache write | Not listed | $2 |
| 1M input tokens, 1-hour cache write | Not listed | $4 |
| 1M input tokens, Highspeed endpoint | Not listed | $4 |
| 1M output tokens, Highspeed endpoint | Not listed | $20 |
Your own bill will differ. These figures use list prices and example workloads. The AI Model Cost Audit reviews your real usage, shortlists models that score as well for each task for less, and estimates the savings.
Launch price $100, regular price $150
Verdict
When to choose each
Hy4
Where it leads
- Lower input and output prices, $0.834 and $2.501 vs $2 and $10 per 1M tokens
- Cheaper cached input, $0.042 vs $0.20 per 1M tokens
Best for
- Agents that work across terminals, desktop apps and browsers on a tight budget
- Finance and office work: research reports, spreadsheets and slides
Watch out for
- API only for now: Tencent says it will publish the weights and a system card later, and only Hy4 preview (August 2026, Apache 2.0) is open today
- In Tencent's own table it scores 44.4% on Terminal-Bench 4.0 against 66.8% for Kimi K3.1 Pro, and Tencent says a clear gap remains to top proprietary models on research-level math, long-horizon coding and hard visual questions
- Tencent publishes no context window or output limit for Hy4 (the preview card lists 1M tokens), and no independent benchmark score is recorded here yet
Kimi K3.1 Pro
Where it leads
- Accepts video input
- Open weights you can run yourself
Best for
- Long-horizon coding and agent work at mid-tier prices
- Long documents, images and video in one 1M-token context
Watch out for
- Thinking is always on: reasoning effort runs low, high or max, and max is the default
- The Highspeed endpoint, two to three times faster by Moonshot's account, costs twice as much at $4 input and $20 output per 1M tokens
- The weights carry Moonshot's own license: a model-as-a-service business above $20M in yearly revenue needs a separate agreement, and products above 100M monthly users or $20M monthly revenue must display the Kimi K3.1 Pro name
- Self-hosting takes a multi-node GPU cluster: the MXFP4 weights alone are about 1.56 TB
- Epoch AI and Arena Text have not scored it yet
From benchmark to production
Building on Hy4 or Kimi K3.1 Pro?
Public scores rarely hold on your own data. We test both models on your real tasks, then build the agent or pipeline around the one that wins.
AI Model Cost AuditLaunch price $100, regular price $150What the audit includes
FAQ
Hy4 vs Kimi K3.1 Pro FAQ
Which is cheaper, Hy4 or Kimi K3.1 Pro?
Hy4 is 58% cheaper per 1M input tokens than Kimi K3.1 Pro, $0.834 vs $2. For 10M input and 2M output tokens, Hy4 costs $13.34 and Kimi K3.1 Pro $40 at standard rates.
Which is cheaper with prompt caching, Hy4 or Kimi K3.1 Pro?
Hy4 charges $0.042 per 1M cached input tokens, 95% below its standard input rate, and Kimi K3.1 Pro $0.20, 90% below. With 9M of 10M input tokens cached and 2M output tokens, Hy4 costs $6.21 and Kimi K3.1 Pro $23.80.
How do Hy4 and Kimi K3.1 Pro compare in Tencent's own results?
In Hy4 launch table (November 4, 2026), Kimi K3.1 Pro leads on DeepSWE v1.1, 76.3% to 74.9%; Kimi K3.1 Pro leads on SWE-Marathon v1.1, 53.8% to 43.8%; Hy4 leads on ProgramBench (almost solved), 30.0% to 24.5%. These are Tencent's figures, not independent measurements.
How do Hy4 and Kimi K3.1 Pro differ in features?
Hy4 takes text and image input; Kimi K3.1 Pro takes text, image and video.
Can I run Hy4 or Kimi K3.1 Pro myself?
Kimi K3.1 Pro is released as open weights you can run yourself; Hy4 is available through its provider's API only.
Further reading
More on Hy4 and Kimi K3.1 Pro
Articles on Hy4 and Kimi K3.1 Pro
- Kimi K3.1 Pro: Benchmarks, Price and Should You Switch?October 25, 2026
Related comparisons
- Claude Opus 5.5 vs Hy4
- Claude Fable 5.1 vs Kimi K3.1 Pro
- DeepSeek-V4.1-Flash vs Hy4
- Claude Opus 5.5 vs Kimi K3.1 Pro
- Gemini 3.8 Flash vs Hy4
- Claude Sonnet 5.5 vs Kimi K3.1 Pro
- GLM-5.3-Flash vs Hy4
- Claude Sonnet 5 vs Kimi K3.1 Pro
- GLM-5.3 vs Hy4
- DeepSeek-V4.1-Flash vs Kimi K3.1 Pro
- GPT-6.1 Sol vs Hy4
- Gemini 4 Argon vs Kimi K3.1 Pro
Production evaluation
Choosing between Hy4 and Kimi K3.1 Pro?
Public benchmarks narrow the shortlist. We run both on your own tasks, data and cost limits, then build the routing and review controls to run the winner in production.
- A head-to-head evaluation on your real workload
- Routing between models by cost and difficulty