Coding work
Code generation, repair, and software-engineering tasks
Not enough matched evidence
Hy4 is not ranked on the public lane for coding, so no winner is named for coding.
Confidence: limited
Hy4 vs Hy4 preview
Updated November 4, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty. This is a same-family comparison, so migration details appear when the source data supports them.
Share or export
Decision reading
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. 8 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
- Shared results
- 8
- Hy4 only
- 7
- Hy4 preview only
- 25
- Like-for-like categories
- 0 / 8
Estimated: Hy4 preview. Conditional ranges do not establish rank confidence.How the comparison works
Which one for your work
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
No workload recommendation clears the current evidence threshold.
Use the matched evidence, workload costs, and sourced specifications below instead of treating a point score as a universal answer.
Show secondary and unsupported calls
Agentic work
Tool use, computer use, and multi-step task completion
Not enough matched evidence
Hy4 is not ranked on the public lane for agentic, so no winner is named for agentic.
Confidence: limitedLong documents
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Confidence: documentedChat turn cost
1K fresh input + 500 output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-ratesCache-heavy agent loop cost
200K cached + 20K fresh input + 10K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-ratesRepository review cost
50K fresh input + 3K output tokens
Not enough matched evidence
A complete comparable API-rate estimate is not available for both models.
Confidence: listed-rates
Which one for a specific job
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
—Hy456.3Hy4 preview
Not comparable · BenchAlign v5.8
The coding row is not comparable on the public lane: at least one model is not measured or not ranked there.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Same basis rules as the category table below
What is actually comparable
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
7 Hy4 only
8 Shared
25 Hy4 preview only
Shape of the matched evidence
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
AutomationBenchAgentic
Normalized gap 24.1Hy4:56.2%Hy4 preview:32.1%HLEKnowledge
Normalized gap 3.7Hy4:59.1%Hy4 preview:55.4%
Bars run 0–100 on each benchmark’s normalized display scale
Category results, on a stated basis
Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
Agentic
Not comparable- Hy4
- Not ranked
- Hy4 preview
- 57.3
- Estimated · #37/135
- Basis
- BenchAlign v5.8 lane · 7 vs 17 public rows
- Reading
- Not comparable
Coding
Not comparable- Hy4
- Not ranked
- Hy4 preview
- 56.3
- Supported · #26/156
- Basis
- BenchAlign v5.8 lane · 5 vs 8 public rows
- Reading
- Not comparable
Reasoning
Not comparable- Hy4
- Not ranked
- Hy4 preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Multimodal
Not comparable- Hy4
- Not ranked
- Hy4 preview
- 86.6
- Unranked · 1 rankable row
- Basis
- Provisional lane · 1 vs 1 weighted rows
- Reading
- Not comparable
Knowledge
Not comparable- Hy4
- Not ranked
- Hy4 preview
- 64.2
- Estimated · #37/179
- Basis
- BenchAlign v5.8 lane · 1 vs 4 public rows
- Reading
- Not comparable
Multilingual
Not comparable- Hy4
- Not ranked
- Hy4 preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Instruction following
Not comparable- Hy4
- Not ranked
- Hy4 preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
Math
Not comparable- Hy4
- Not ranked
- Hy4 preview
- Not ranked
- Basis
- Provisional lane · 0 vs 0 weighted rows
- Reading
- Not comparable
| Category | Hy4 | Hy4 preview | Basis | Reading |
|---|---|---|---|---|
| Agentic | Not ranked | 57.3Estimated · #37/135 | Not comparableBenchAlign v5.8 lane · 7 vs 17 public rows | Not comparable |
| Coding | Not ranked | 56.3Supported · #26/156 | Not comparableBenchAlign v5.8 lane · 5 vs 8 public rows | Not comparable |
| Reasoning | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multimodal | Not ranked | 86.6Unranked · 1 rankable row | Not comparableProvisional lane · 1 vs 1 weighted rows | Not comparable |
| Knowledge | Not ranked | 64.2Estimated · #37/179 | Not comparableBenchAlign v5.8 lane · 1 vs 4 public rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Supported evidence per lane · bars run 0–100Methodology
What each workload costs
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
Chat turn
1K fresh input + 500 output tokens
- Hy4
- $0.00208
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
Hy4 preview has no comparable published API token rate.
Repository review
50K fresh input + 3K output tokens
- Hy4
- $0.0492
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
Hy4 preview has no comparable published API token rate.
Cache-heavy agent loop
200K cached + 20K fresh input + 10K output tokens
- Hy4
- $0.05009
- Fits in one request
- Hy4 preview
- Self-hosted; infrastructure cost varies
- Fits in one request
- Cached-input rate unavailable
Hy4 preview has no comparable published API token rate.
Cached input falls back to the list input rate only where a cached rate is unpublished
Specification differences
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
SpecificationHy4Hy4 preview
Context window
Maximum documented context; output-token limits may be lower.
Hy4 preview
API model ID
Hy4
tencent/hy4
Tencent Cloud TokenHub model pricingHy4 preview
tencent/Hy4-preview
Tencent Hy4 preview model cardCached-input rate
A missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
Hy4
$0.042 per 1M cached input tokens
Tencent Cloud TokenHub model pricingHy4 preview
No comparable hosted API rate
Tencent Hy4 preview model cardDocumented inputs
Hy4
Not sourced
Hy4 preview
Not sourced
Documented outputs
Hy4
Not sourced
Hy4 preview
Not sourced
Provider availability
Hy4
Not sourced
Hy4 preview
Not sourced
Reasoning profile
Hy4
Reasoning
Hy4 preview
Reasoning
Weight access
Hy4
Pending
Hy4 preview
Open Weight
License
Hy4
Pending
Hy4 preview
Open Weight
Release date
Hy4
2026-11-04
Hy4 preview
2026-08-28
If you are considering the documented upgrade path
- Deployment change
- Both entries list Tencent as the provider. Confirm endpoint, model ID, limits, and feature support before switching.
- Quality signal
- At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner.
- Workload cost
- A complete comparable API-rate estimate is not available for both models.
- Context tradeoff
- Both models list 1M.
Run the same representative tasks against both endpoints before changing production traffic.
Questions
Which is better, Hy4 or Hy4 preview?
At least one model is not scored in the current public ranking lane, so the page does not name an overall quality winner. The page therefore keeps the decision tied to the specific documented workload.
Which is better for coding, Hy4 or Hy4 preview?
Hy4 is not ranked on the public lane for coding, so no winner is named for coding.
Which is better for agentic tasks, Hy4 or Hy4 preview?
Hy4 is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.
Which costs less, Hy4 or Hy4 preview?
Both models do not have comparable published API token rates, so this page does not name a universal price winner.
Which has the larger context window, Hy4 or Hy4 preview?
Both models list the same context window, 1M.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Hy4
API / mo$2,501
Self-host / moNot listed
Break-even—
Proprietary model — self-hosting not applicable.
Hy4 preview
API / mo$0
Self-host / mo$27,962
Break-even—
Benchmark evidence
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
Browse raw public benchmark evidence40 rows
Agentic
Terminal-Bench 4.0
Hy444.40%SourceHy4 preview—Not directly comparable
Terminal-Bench-Science 0.1
Hy419.8%SourceHy4 preview—Not directly comparable
HLE w/ tools
Hy459.1%SourceHy4 preview55.4%SourceHy4 leads this result
Agents' Last Exam
Hy441.0%SourceHy4 preview22.8%SourceHy4 leads this result
APEX-Agents
Hy475.2%SourceHy4 preview37.1%SourceHy4 leads this result
AutomationBench
Hy456.2%SourceHy4 preview32.1%SourceHy4 leads this result
OSWorld 2.1
Hy478.9%SourceHy4 preview—Not directly comparable
Terminal-Bench 2.1
Hy4—Hy4 preview85.4%SourceNot directly comparable
CyberGym
Hy4—Hy4 preview78.4%SourceNot directly comparable
WideResearch
Hy4—Hy4 preview83.9%SourceNot directly comparable
DRACO
Hy4—Hy4 preview77.2%SourceNot directly comparable
MCP Atlas
Hy4—Hy4 preview83.7%SourceNot directly comparable
Toolathlon-Verified
Hy4—Hy4 preview74.1%SourceNot directly comparable
skillsBench
Hy4—Hy4 preview62.9%SourceNot directly comparable
JobBench
Hy4—Hy4 preview61.7%SourceNot directly comparable
BankerToolBench
Hy4—Hy4 preview78.6%SourceNot directly comparable
Terminal-Bench 2.1 (Vals)
Hy4—Hy4 preview55.1%SourceNot directly comparable
CWE-bench v1
Hy4—Hy4 preview53.0%SourceNot directly comparable
Hermes Index
Hy4—Hy4 preview26.1%SourceNot directly comparable
Hermes Bench
Hy4—Hy4 preview58.6%SourceNot directly comparable
Coding
DeepSWE
Hy474.9%SourceHy4 preview64.3%SourceHy4 leads this result
sweMarathon
Hy443.8%SourceHy4 preview31.9%SourceHy4 leads this result
ProgramBench
Hy430.0%SourceHy4 preview17.5%SourceHy4 leads this result
FrontierCode 2.0
Hy431.0%SourceHy4 preview—Not directly comparable
PostTrainBench v1.2
Hy435.2%SourceHy4 preview—Not directly comparable
Terminal-Bench 2.1
Hy4—Hy4 preview85.4%SourceNot directly comparable
SWE-bench Pro
Hy4—Hy4 preview65.7%SourceNot directly comparable
SWE Multilingual
Hy4—Hy4 preview82.9%SourceNot directly comparable
NL2Repo
Hy4—Hy4 preview58.9%SourceNot directly comparable
PostTrain Bench
Hy4—Hy4 preview35.6%SourceNot directly comparable
Reasoning
Multimodal
Knowledge
Math
Apex
Hy4—Hy4 preview74.2%SourceNot directly comparable
40 public results · 8 shared
Watch Hy4 vs Hy4 preview
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 6,100+ subscribers.
Last updated November 4, 2026