Fugu Ultra vs GPT-6 Astra vs Claude Mythos 5
Too close to call on our weighted score (GPT-6 Astra 36, Claude Mythos 5 36, Fugu Ultra 33). The right pick depends on what you value most.
Sakana AI
Fugu Ultra
33/100- ECI—
- Price$5.00 / $30.00
- Context1M
OpenAIGPT-6 Astra
36/100- ECI166.5
- Price$10.00 / $50.00
- Context1.05M
AnthropicClaude Mythos 5
36/100- ECI—
- Price$10.00 / $50.00
- Context1M
Too close to call
It is close. Our weighted score puts them within a point (GPT-6 Astra 36/100, Claude Mythos 5 36/100, Fugu Ultra 33/100), so choose by what matters most for your work: Fugu Ultra on price and GPT-6 Astra for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. 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 priceFugu UltraFugu Ultra $11.25 · GPT-6 Astra $20.00 · Claude Mythos 5 $20.00 per 1M tokens (3:1 blend)
- Longest contextGPT-6 AstraGPT-6 Astra 1,050,000 · Fugu Ultra 1,000,000 · Claude Mythos 5 1,000,000 tokens
- Widest inputsGPT-6 Astra and Claude Mythos 5Fugu Ultra: Text, Images · GPT-6 Astra: Text, Images, PDFs · Claude Mythos 5: Text, Images, PDFs
- Self-hostingNo open weightsAll three are available only through APIs
| Measure | Weight | Fugu Ultra | GPT-6 Astra | Claude Mythos 5 |
|---|---|---|---|---|
| Price | 50% | 0 | 0 | 0 |
| Inputs & features | 30% | 70 | 80 | 80 |
| Context window | 20% | 60 | 61 | 60 |
| Overall | 100% | 33/100 | 36/100 | 36/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Fugu UltraSakana AI | GPT-6 AstraOpenAI | Claude Mythos 5Anthropic |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | 166.5 | — |
| ECI rank | — | #2 of 148 | — |
| GPQA DiamondGraduate-level science questions | — | 95.8% | — |
| FrontierMath Tiers 1–3Research-level mathematics | — | 93.7% | — |
| OTIS Mock AIME 2024–2025Competition mathematics | — | 100% | — |
| SimpleQA VerifiedShort factual questions | — | 75.6% | — |
| Price per million tokens | |||
| Input | $5.00 (best) | $10.00 | $10.00 |
| Output | $30.00 (best) | $50.00 | $50.00 |
| Cached input | $0.50 (best) | $1.00 | — |
| Blended (3:1) | $11.25 (best) | $20.00 | $20.00 |
| Long-context rate | Over 272K: $10.00 / $45.00 | Over 272K: $20.00 / $75.00 | Same rate |
| Price source | Official Sakana AI API | Official OpenAI API | Median of 2 providers |
| Limits | |||
| Context window | 1,000,000 tokens | 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 | No | Yes | Yes |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yeshigh · xhigh | Yeslow · medium · high · xhigh · max | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | Yes | Yes |
| Availability | |||
| Weights | Proprietary | Proprietary | Proprietary |
| API model ID | fugu-ultra | gpt-6-astra | — |
| API providers | 11 | 26 (best) | 2 |
| Released | Jun 15, 2026 | Sep 4, 2026 | Jun 9, 2026 |
| Knowledge cutoff | — | Apr 30, 2026 | Jan 31, 2026 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Fugu Ultra$110.00
- GPT-6 Astra$200.00
- Claude Mythos 5$200.00
Which should you choose?
Which is better: Fugu Ultra, GPT-6 Astra or Claude Mythos 5?
It is close. Our weighted score puts them within a point (GPT-6 Astra 36/100, Claude Mythos 5 36/100, Fugu Ultra 33/100), so choose by what matters most for your work: Fugu Ultra on price and GPT-6 Astra for long inputs. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Fugu Ultra, GPT-6 Astra or Claude Mythos 5?
Fugu Ultra is cheaper at $5.00 input / $30.00 output per million tokens (official Sakana AI API price). GPT-6 Astra costs $10.00 input / $50.00 output per million tokens (official OpenAI API price); Claude Mythos 5 costs $10.00 input / $50.00 output per million tokens (median across 2 API providers). At a typical mix of three input tokens to one output token, that is $11.25 per million tokens for Fugu Ultra versus $20.00 for GPT-6 Astra (1.8× as much) and $20.00 for Claude Mythos 5 (1.8× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Fugu Ultra has not been scored yet, GPT-6 Astra has an ECI of 166.5 and Claude Mythos 5 has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Fugu Ultra, GPT-6 Astra and Claude Mythos 5 yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
GPT-6 Astra has the largest context window at 1,050,000 tokens, against 1,000,000 for Fugu Ultra and 1,000,000 for Claude Mythos 5. Maximum output per response: GPT-6 Astra up to 128,000, Claude Mythos 5 up to 128,000 tokens.
Which can read images, PDFs, audio or video?
Fugu Ultra accepts text and images; GPT-6 Astra accepts text, images and PDFs; Claude Mythos 5 accepts text, images and PDFs. GPT-6 Astra handles the widest range of inputs.
Are any of these open source?
No. Fugu Ultra, GPT-6 Astra and Claude Mythos 5 are proprietary and only available through APIs and apps.
Which is newer?
GPT-6 Astra is the newest, released Sep 4, 2026. Fugu Ultra came out Jun 15, 2026; Claude Mythos 5 came out Jun 9, 2026. Knowledge cutoff: GPT-6 Astra Apr 30, 2026, Claude Mythos 5 Jan 31, 2026.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, 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 Oct 4, 2026.
