GPT-6.2 Astra vs Claude Mythos 5.2 vs Claude Fable 5.2
No independent capability benchmark covers every model here yet. The score uses price.
Price in rating: Included · Exclude price
OpenAI
GPT-6.2 Astra
0/100- ECI174.2
- Price$8.00 / $40.00
- Context1.05M
Anthropic
Claude Mythos 5.2
0/100- ECI—
- Price$8.00 / $40.00
- Context—
Anthropic
Claude Fable 5.2
0/100- ECI173.5
- Price$8.00 / $40.00
- Context1M
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (GPT-6.2 Astra 0/100, Claude Mythos 5.2 0/100, Claude Fable 5.2 0/100) compare the available specifications and price. The score weighs price 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceSame priceGPT-6.2 Astra $16.00 · Claude Mythos 5.2 $16.00 · Claude Fable 5.2 $16.00 per 1M tokens (3:1 blend)
- Longest contextGPT-6.2 AstraGPT-6.2 Astra 1,050,000 · Claude Fable 5.2 1,000,000 tokens
- Widest inputsGPT-6.2 Astra and Claude Fable 5.2GPT-6.2 Astra: Text, Images, PDFs · Claude Mythos 5.2: Text, Images · Claude Fable 5.2: Text, Images, PDFs
- Self-hostingNo confirmed open weightsAll three are available only through APIs
| Measure | Weight | GPT-6.2 Astra | Claude Mythos 5.2 | Claude Fable 5.2 |
|---|---|---|---|---|
| Price | 100% | 0 | 0 | 0 |
| Specs & price only | 100% | 0/100 | 0/100 | 0/100 |
Left out because at least one model lacks the data: capability, inputs & features and context window. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Task | — | Restricted-access reasoning | — |
| Published capabilities | A frontier reasoning model for coding, computer use, and professional work such as slides, spreadsheets, and documents. | Shares its underlying model with Claude Fable 5.2. Access and safeguards differ. | Claude model for demanding reasoning and long-horizon agentic work |
| Access | — | Vetted organizations through Anthropic verification programs | — |
| Sources | — | Publisher documentation | — |
| Capability | |||
| Capabilities Index (ECI) | 174.2 (best) | — | 173.5 |
| ECI rank | #1 of 157 (best) | — | #2 of 157 |
| GPQA DiamondGraduate-level science questions | 96.8% (best) | — | 96.0% |
| FrontierMath Tiers 1–3Research-level mathematics | 92.6% | — | 95.1% (best) |
| OTIS Mock AIME 2024–2025Competition mathematics | 100% (best) | — | 99.7% |
| SimpleQA VerifiedShort factual questions | 78.5% | — | 94.0% (best) |
| Price per million tokens | |||
| Input | $8.00 | $8.00 | $8.00 |
| Output | $40.00 | $40.00 | $40.00 |
| Cached input | $0.40 | — | $0.20 (best) |
| Blended (3:1) | $16.00 | $16.00 | $16.00 |
| Long-context rate | Over 272K: $16.00 / $60.00 | Same rate | Same rate |
| Price source | Official OpenAI API | Official Anthropic API | Official Anthropic API |
| Limits | |||
| Context window | 1,050,000 tokens (best) | — | 1,000,000 tokens |
| Max output | 128,000 tokens | — | 128,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | Yes |
| PDFs | Yes | No | Yes |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yeslow · medium · high · xhigh · max | Unconfirmed | Yeslow · medium · high · xhigh · max |
| Tool calling | Yes | — | Yes |
| Structured output | Yes | — | Yes |
| Availability | |||
| Weights | Proprietary | Proprietary | Proprietary |
| API model ID | gpt-6.2-astra | — | claude-fable-5-2 |
| API providers | 22 | — | 26 (best) |
| Released | Nov 17, 2026 | Nov 9, 2026 | Nov 9, 2026 |
| Knowledge cutoff | May 18, 2026 | — | Jun 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
GPT-6.2 Astra$160.00
Claude Mythos 5.2$160.00
Claude Fable 5.2$160.00
Which should you choose?
Which is better: GPT-6.2 Astra, Claude Mythos 5.2 or Claude Fable 5.2?
There is not enough shared capability evidence to name an overall winner. The scores (GPT-6.2 Astra 0/100, Claude Mythos 5.2 0/100, Claude Fable 5.2 0/100) compare the available specifications and price. The score weighs price 100%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, GPT-6.2 Astra, Claude Mythos 5.2 or Claude Fable 5.2?
GPT-6.2 Astra, Claude Mythos 5.2 and Claude Fable 5.2 cost the same: $8.00 input / $40.00 output per million tokens.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. GPT-6.2 Astra has an ECI of 174.2, Claude Mythos 5.2 has not been scored yet and Claude Fable 5.2 has an ECI of 173.5.
Which is better for coding?
There are no published SWE-bench Verified results for GPT-6.2 Astra, Claude Mythos 5.2 and Claude Fable 5.2 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. Note that Claude Mythos 5.2 has no confirmed tool-calling support in this comparison.
Which has the bigger context window?
Published context limits: GPT-6.2 Astra 1,050,000 tokens, Claude Mythos 5.2 unconfirmed, Claude Fable 5.2 1,000,000 tokens. Maximum output per response: GPT-6.2 Astra up to 128,000, Claude Fable 5.2 up to 128,000 tokens.
Which can read images, PDFs, audio or video?
GPT-6.2 Astra accepts text, images and PDFs; Claude Mythos 5.2 accepts text and images; Claude Fable 5.2 accepts text, images and PDFs. GPT-6.2 Astra handles the widest range of inputs.
Are any of these open source?
No. GPT-6.2 Astra, Claude Mythos 5.2 and Claude Fable 5.2 are proprietary and only available through APIs and apps.
Which is newer?
GPT-6.2 Astra is the newest, released Nov 17, 2026. Claude Mythos 5.2 came out Nov 9, 2026; Claude Fable 5.2 came out Nov 9, 2026. Knowledge cutoff: GPT-6.2 Astra May 18, 2026, Claude Fable 5.2 Jun 2026.
How do you decide the winner?
This rating uses price (100%, blended price per million tokens (3 input : 1 output), log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 19, 2026.