Skip to main content
BenchLM
Data

GPT-6 Astra vs Hy4 flash

Updated November 17, 2026. We do not rank this pair: at least one has no public score. Public scores include evidence status and uncertainty.

Share or export
Share on XLinkedInAPI/MCP

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. 5 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.

Model A
OpenAI logo

OpenAI

85.62/100

Supported · Public rank #4

90% interval 80.3–91.0

Model B
Tencent logo

Tencent

—

Evidence status unavailable

90% interval unavailable

API price per 1M tokens

GPT-6 Astra: $10 input / $50 output

Hy4 flash: $0.070 input / $0.28 output

Context window

GPT-6 Astra: 1.05M

Hy4 flash: 1M

Released

GPT-6 Astra: September 3, 2026

Hy4 flash: November 16, 2026

Shared results
5
GPT-6 Astra only
53
Hy4 flash only
5
Like-for-like categories
0 / 8
Supported: GPT-6 AstraHow 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.

  • Long documents

    Prompts that approach the documented context limit

    GPT-6 Astra

    GPT-6 Astra has the larger documented context window.

    Confidence: documented
  • Chat turn cost

    1K fresh input + 500 output tokens

    Hy4 flash

    Hy4 flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Cache-heavy agent loop cost

    200K cached + 20K fresh input + 10K output tokens

    Hy4 flash

    Hy4 flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
Show secondary and unsupported calls
  • Repository review cost

    50K fresh input + 3K output tokens

    Hy4 flash

    Hy4 flash has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.

    Confidence: listed-rates
  • Coding work

    Code generation, repair, and software-engineering tasks

    Not enough matched evidence

    Hy4 flash is not ranked on the public lane for coding, so no winner is named for coding.

    Confidence: limited
  • Agentic work

    Tool use, computer use, and multi-step task completion

    Not enough matched evidence

    Hy4 flash is not ranked on the public lane for agentic, so no winner is named for agentic.

    Confidence: limited

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.

78.9GPT-6 Astra—Hy4 flash

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

Or describe your own job

The LLM Selector reads what you need a model for and returns a shortlist, ranked in the same order the job pages publish.

Describe the task and constraints. This selector recommends models; it cannot carry out the task or access your files and accounts.

We keep what you type here, drafts included, for 180 days, and after that only the job and requirements read from it. Nothing is tied to your IP address or account. Leave out names and other personal details. Privacy

What is actually comparable

Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.

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.

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, reasoning, multimodal, 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
GPT-6 Astra
73.8
Supported · #9/138
Hy4 flash
Not ranked
Basis
BenchAlign v5.8 lane · 20 vs 5 public rows
Reading
Not comparable

Coding

Not comparable
GPT-6 Astra
78.9
Supported · #5/163
Hy4 flash
Not ranked
Basis
BenchAlign v5.8 lane · 10 vs 2 public rows
Reading
Not comparable

Reasoning

Not comparable
GPT-6 Astra
81.3
Supported · #3/160
Hy4 flash
Not ranked
Basis
BenchAlign v5.8 lane · 10 vs 0 public rows
Reading
Not comparable

Multimodal

Not comparable
GPT-6 Astra
78.8
Supported · #3/67
Hy4 flash
Not ranked
Basis
BenchAlign v5.8 lane · 6 vs 2 public rows
Reading
Not comparable

Knowledge

Not comparable
GPT-6 Astra
85.7
Supported · #4/181
Hy4 flash
Not ranked
Basis
BenchAlign v5.8 lane · 9 vs 1 public rows
Reading
Not comparable

Multilingual

Not comparable
GPT-6 Astra
Not ranked
Hy4 flash
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Instruction following

Not comparable
GPT-6 Astra
Not ranked
Hy4 flash
Not ranked
Basis
Provisional lane · 0 vs 0 weighted rows
Reading
Not comparable

Math

Not comparable
GPT-6 Astra
93.8
Unranked · 2 rankable rows
Hy4 flash
Not ranked
Basis
Provisional lane · 2 vs 0 weighted rows
Reading
Not comparable

Ranks count the models scored in each category’s lane, so the agentic, coding, reasoning, multimodal, 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

GPT-6 Astra
$0.035
Fits in one request
Hy4 flash
$0.00021
Fits in one request

Hy4 flash has the lower modeled cost

Costs use the listed standard API rates.

Repository review

50K fresh input + 3K output tokens

GPT-6 Astra
$0.65
Fits in one request
Hy4 flash
$0.00434
Fits in one request

Hy4 flash has the lower modeled cost

Costs use the listed standard API rates.

Cache-heavy agent loop

200K cached + 20K fresh input + 10K output tokens

GPT-6 Astra
$0.9
Fits in one request
Hy4 flash
$0.00481
Fits in one request

Hy4 flash has the lower modeled cost

Costs use the listed standard API rates.

Your workload

Enter a month of tokens to price it at both models’ listed API rates.

GPT-6 Astra
$200.00
per month
Hy4 flash
$1.26
per month

Standard listed rates, no batch or volume discounts.

Cached input falls back to the list input rate only where a cached rate is unpublished

Price against score, among all priced models

Overall score against output price per 1M tokens, one model per family. Pricing as of November 17, 2026.

GPT-6 Astra

20406080100$0.10$0.50$1$5$10$50$100Output price, $ per 1M tokens (log)Overall score
  • 3 cheaper models score at least as high as GPT-6 Astra. The nearest is Claude Fable 5.2 (score 93.9, $40.00 per 1M output tokens).
  • Hy4 flash has no overall score yet, so it is not plotted.

Gray dots are the other priced models. The dark step line is the efficiency frontier: for a model under it, another model costs no more and scores at least as high.

See GPT-6 Astra on the full price vs performance chart

Specification differences

Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.

Provider availability

GPT-6 Astra

Limited Availability · OpenAI Responses API, OpenAI Chat Completions API, ChatGPT Plus, Pro, Business, and Enterprise, Microsoft Azure, AWS Bedrock

OpenAI GPT-6 Astra model documentation

Hy4 flash

Not sourced

Reasoning profile

GPT-6 Astra

Reasoning

Hy4 flash

Reasoning

Weight access

GPT-6 Astra

Proprietary

Hy4 flash

Pending

License

GPT-6 Astra

Proprietary

Hy4 flash

Pending

Release date

GPT-6 Astra

2026-09-03

Hy4 flash

2026-11-16

If you already use one of these models

Deployment change
The models list different providers, so authentication, endpoint behavior, limits, and feature support may change.
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
Repository review: $0.65 vs $0.00434. Cache-heavy agent loop: $0.9 vs $0.00481.
Context tradeoff
GPT-6 Astra has the larger documented window (1.05M).
Run the same representative tasks against both endpoints before changing production traffic.

Questions

Which is better, GPT-6 Astra or Hy4 flash?

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, GPT-6 Astra or Hy4 flash?

Hy4 flash is not ranked on the public lane for coding, so no winner is named for coding.

Which is better for agentic tasks, GPT-6 Astra or Hy4 flash?

Hy4 flash is not ranked on the public lane for agentic tasks, so no winner is named for agentic tasks.

Which costs less, GPT-6 Astra or Hy4 flash?

For the stated presets, chat costs $0.035 on GPT-6 Astra and $0.00021 on Hy4 flash; repository review costs $0.65 and $0.00434; the cache-heavy agent loop costs $0.9 and $0.00481. Costs use the listed standard API rates.

Which has the larger context window, GPT-6 Astra or Hy4 flash?

GPT-6 Astra has the larger documented context window: 1.05M, compared with 1M.

Benchmark evidence

The full public result ledger is available for audit without forcing a wide desktop table onto a phone.

Browse raw public benchmark evidence63 rows

Agentic

  • Terminal-Bench 4.1

    GPT-6 Astra60.30%
    Source
    Hy4 flash—

    Not directly comparable

  • LifeVal

    GPT-6 Astra80.6%
    Source
    Hy4 flash—

    Not directly comparable

  • BrowseComp

    GPT-6 Astra91.5%
    Source
    Hy4 flash—

    Not directly comparable

  • OSWorld 2.0

    GPT-6 Astra72.6%
    Source
    Hy4 flash—

    Not directly comparable

  • Terminal-Bench 4.0

    GPT-6 Astra57.90%
    Source
    Hy4 flash19.20%
    Source

    GPT-6 Astra leads this result

  • Terminal-Bench-Science 0.1

    GPT-6 Astra64.6%
    Source
    Hy4 flash—

    Not directly comparable

  • ExploitGym

    GPT-6 Astra42.4%
    Source
    Hy4 flash—

    Not directly comparable

  • Agents' Last Exam

    GPT-6 Astra59.3%
    Source
    Hy4 flash—

    Not directly comparable

  • AutomationBench

    GPT-6 Astra41.4%
    Source
    Hy4 flash52.2%
    Source

    Hy4 flash leads this result

  • HLE w/ tools

    GPT-6 Astra57.2%
    Source
    Hy4 flash55.2%
    Source

    GPT-6 Astra leads this result

  • Terminal-Bench 2.1 (Vals)

    GPT-6 Astra87.3%
    Source
    Hy4 flash—

    Not directly comparable

  • ApprenticeBench

    GPT-6 Astra68%
    Source
    Hy4 flash—

    Not directly comparable

  • CWE-bench v1

    GPT-6 Astra68.0%
    Source
    Hy4 flash—

    Not directly comparable

  • Hermes Index

    GPT-6 Astra56.3%
    Source
    Hy4 flash—

    Not directly comparable

  • Hermes Bench

    GPT-6 Astra73.7%
    Source
    Hy4 flash—

    Not directly comparable

  • MazeBench

    GPT-6 Astra24%
    Source
    Hy4 flash—

    Not directly comparable

  • OSWorld 2.1

    GPT-6 Astra75.8%
    Source
    Hy4 flash68.1%
    Source

    GPT-6 Astra leads this result

  • RoboDojo Sim

    GPT-6 Astra29.8%
    Source
    Hy4 flash—

    Not directly comparable

  • Remote Labor Index

    GPT-6 Astra20.8%
    Source
    Hy4 flash—

    Not directly comparable

  • CWE-bench

    GPT-6 Astra64.2%
    Source
    Hy4 flash—

    Not directly comparable

  • APEX-Agents

    GPT-6 Astra—
    Hy4 flash63.1%
    Source

    Not directly comparable

Coding

  • ScarfBench

    GPT-6 Astra65.4%
    Source
    Hy4 flash—

    Not directly comparable

  • DeepSWE

    GPT-6 Astra74.1%
    Source
    Hy4 flash64.6%
    Source

    GPT-6 Astra leads this result

  • FrontierCode 1.1 Main

    GPT-6 Astra53.3%
    Source
    Hy4 flash—

    Not directly comparable

  • FrontierCode 1.1 Extended

    GPT-6 Astra64.5%
    Source
    Hy4 flash—

    Not directly comparable

  • FrontierSWE v2

    GPT-6 Astra65.5%
    Source
    Hy4 flash—

    Not directly comparable

  • PostTrainBench v1.1

    GPT-6 Astra44.3%
    Source
    Hy4 flash—

    Not directly comparable

  • Bug Hunt Bench

    GPT-6 Astra45.0 fixes
    Source
    Hy4 flash—

    Not directly comparable

  • FrontierCode 2.0

    GPT-6 Astra44.0%
    Source
    Hy4 flash—

    Not directly comparable

  • VulcanBench Frontier v4

    GPT-6 Astra89.3%
    Source
    Hy4 flash—

    Not directly comparable

  • PostTrainBench

    GPT-6 Astra41.9%
    Source
    Hy4 flash—

    Not directly comparable

  • ProgramBench

    GPT-6 Astra—
    Hy4 flash16.0%
    Source

    Not directly comparable

Reasoning

  • BioBasicsBench

    GPT-6 Astra67.0%
    Source
    Hy4 flash—

    Not directly comparable

  • STEMBench

    GPT-6 Astra2.81
    Source
    Hy4 flash—

    Not directly comparable

  • ARC-AGI-1

    GPT-6 Astra97.50%
    Source
    Hy4 flash—

    Not directly comparable

  • ARC-AGI-2

    GPT-6 Astra95%
    Source
    Hy4 flash—

    Not directly comparable

  • ARC-AGI-3

    GPT-6 Astra62.7%
    Source
    Hy4 flash—

    Not directly comparable

  • GeneBench-Pro

    GPT-6 Astra37.8%
    Source
    Hy4 flash—

    Not directly comparable

  • MRCR v2 256K-512K

    GPT-6 Astra100.0%
    Source
    Hy4 flash—

    Not directly comparable

  • MRCR v2 512K-1M

    GPT-6 Astra96.3%
    Source
    Hy4 flash—

    Not directly comparable

  • GraphWalks BFS 256K–1M

    GPT-6 Astra71.8%
    Source
    Hy4 flash—

    Not directly comparable

  • ARC-AGI-3 Provider Adapter

    GPT-6 Astra98.55%
    Source
    Hy4 flash—

    Not directly comparable

Multimodal

  • ExtractBench Pro

    GPT-6 Astra58%
    Source
    Hy4 flash—

    Not directly comparable

  • ExtractBench (LlamaIndex)

    GPT-6 Astra91.91%
    Source
    Hy4 flash—

    Not directly comparable

  • ParseBench-1

    GPT-6 Astra69.81%
    Source
    Hy4 flash—

    Not directly comparable

  • ScreenSpot Pro

    GPT-6 Astra92.7%
    Source
    Hy4 flash—

    Not directly comparable

  • BenchCAD Vision2Code (tools)

    GPT-6 Astra0.959
    Source
    Hy4 flash—

    Not directly comparable

  • Biomedical image analysis

    GPT-6 Astra77.5%
    Source
    Hy4 flash—

    Not directly comparable

  • MMMU-Pro

    GPT-6 Astra—
    Hy4 flash75.8%
    Source

    Not directly comparable

  • ZeroBench w/ Python

    GPT-6 Astra—
    Hy4 flash45.8%
    Source

    Not directly comparable

Knowledge

  • GPQA

    GPT-6 Astra96%
    Source
    Hy4 flash—

    Not directly comparable

  • GPQA-D

    GPT-6 Astra96.0%
    Source
    Hy4 flash—

    Not directly comparable

  • HealthBench (raw)

    GPT-6 Astra56.9%
    Source
    Hy4 flash—

    Not directly comparable

  • HealthBench (length-adjusted)

    GPT-6 Astra58.3%
    Source
    Hy4 flash—

    Not directly comparable

  • HealthBench Professional

    GPT-6 Astra64.7%
    Source
    Hy4 flash—

    Not directly comparable

  • HealthBench Professional (raw)

    GPT-6 Astra68.2%
    Source
    Hy4 flash—

    Not directly comparable

  • HealthBench Hard

    GPT-6 Astra36.6%
    Source
    Hy4 flash—

    Not directly comparable

  • PhysicianBench

    GPT-6 Astra48.0%
    Source
    Hy4 flash—

    Not directly comparable

  • Morphology-to-molecule matching

    GPT-6 Astra22.8%
    Source
    Hy4 flash—

    Not directly comparable

  • HLE

    GPT-6 Astra—
    Hy4 flash55.2%
    Source

    Not directly comparable

Instruction following

  • Gray Swan IPI (15 attempts)

    GPT-6 Astra8.5%
    Source
    Hy4 flash—

    Not directly comparable

Math

  • FrontierMath v2 (Tier 4)

    GPT-6 Astra97.600%
    Source
    Hy4 flash—

    Not directly comparable

  • FrontierMath v2 (Tiers 1-3)

    GPT-6 Astra93.684%
    Source
    Hy4 flash—

    Not directly comparable

63 public results · 5 shared

Watch GPT-6 Astra vs Hy4 flash

One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.

Read a sample issue

Join 7,000+ subscribers.

Last updated November 17, 2026