GPT-5.6 Sol vs Laguna XS.2 vs Claude Opus 5
Too close to call on our weighted score (GPT-5.6 Sol 72, Claude Opus 5 72, Laguna XS.2 36). The right pick depends on what you value most.
OpenAIGPT-5.6 Sol
72/100- ECI161.8
- Price$4.00 / $20.00
- Context1.05M
PoolsideLaguna XS.2
36/100- ECI—
- Price—
- Context262K
AnthropicClaude Opus 5
72/100- ECI162.9
- Price$5.00 / $25.00
- Context1M
Too close to call
It is close. Our weighted score puts them within a point (GPT-5.6 Sol 72/100, Claude Opus 5 72/100, Laguna XS.2 36/100), so choose by what matters most for your work: GPT-5.6 Sol on price and GPT-5.6 Sol for long inputs. The score weighs inputs & features 60%, context window 40%. 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 priceGPT-5.6 SolGPT-5.6 Sol $8.00 · Claude Opus 5 $10.00 per 1M tokens (3:1 blend) · Laguna XS.2 unpriced
- Longest contextGPT-5.6 SolGPT-5.6 Sol 1,050,000 · Claude Opus 5 1,000,000 · Laguna XS.2 262,144 tokens
- Widest inputsGPT-5.6 Sol and Claude Opus 5GPT-5.6 Sol: Text, Images, PDFs · Laguna XS.2: Text · Claude Opus 5: Text, Images, PDFs
- Self-hostingLaguna XS.2Publishes downloadable weights
| Measure | Weight | GPT-5.6 Sol | Laguna XS.2 | Claude Opus 5 |
|---|---|---|---|---|
| Inputs & features | 60% | 80 | 35 | 80 |
| Context window | 40% | 61 | 37 | 60 |
| Overall | 100% | 72/100 | 36/100 | 72/100 |
Left out because at least one model lacks the data: capability and price. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | GPT-5.6 SolOpenAI | Laguna XS.2Poolside | Claude Opus 5Anthropic |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | 161.8 | — | 162.9 (best) |
| ECI rank | #8 of 148 | — | #5 of 148 (best) |
| GPQA DiamondGraduate-level science questions | 93.5% | — | 93.9% (best) |
| FrontierMath Tiers 1–3Research-level mathematics | 89.1% (best) | — | 85.6% |
| OTIS Mock AIME 2024–2025Competition mathematics | 100% (best) | — | 98.9% |
| SimpleQA VerifiedShort factual questions | 69.7% (best) | — | 59.9% |
| Price per million tokens | |||
| Input | $4.00 (best) | — | $5.00 |
| Output | $20.00 (best) | — | $25.00 |
| Cached input | $0.40 (best) | — | $0.50 |
| Blended (3:1) | $8.00 (best) | — | $10.00 |
| Long-context rate | Over 272K: $8.00 / $30.00 | — | Same rate |
| Price source | Official OpenAI API | — | Official Anthropic API |
| Limits | |||
| Context window | 1,050,000 tokens (best) | 262,144 tokens | 1,000,000 tokens |
| Max output | 128,000 tokens (best) | 32,768 tokens | 128,000 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | No | Yes |
| PDFs | Yes | No | Yes |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yeslow · medium · high · xhigh · max | Yes | Yeslow · medium · high · xhigh · max |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | Yes |
| Availability | |||
| Weights | Proprietary | Open | Proprietary |
| API model ID | gpt-5.6-sol | — | claude-opus-5 |
| API providers | 40 (best) | 1 | 35 |
| Released | Jul 9, 2026 | Apr 28, 2026 | Jul 24, 2026 |
| Knowledge cutoff | Feb 16, 2026 | — | May 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-5.6 Sol$80.00
- Laguna XS.2—
- Claude Opus 5$100.00
Which should you choose?
Which is better: GPT-5.6 Sol, Laguna XS.2 or Claude Opus 5?
It is close. Our weighted score puts them within a point (GPT-5.6 Sol 72/100, Claude Opus 5 72/100, Laguna XS.2 36/100), so choose by what matters most for your work: GPT-5.6 Sol on price and GPT-5.6 Sol for long inputs. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, GPT-5.6 Sol, Laguna XS.2 or Claude Opus 5?
GPT-5.6 Sol is cheaper at $4.00 input / $20.00 output per million tokens (official OpenAI API price). Claude Opus 5 costs $5.00 input / $25.00 output per million tokens (official Anthropic API price). At a typical mix of three input tokens to one output token, that is $8.00 per million tokens for GPT-5.6 Sol versus $10.00 for Claude Opus 5 (1.3× as much). Laguna XS.2 has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. GPT-5.6 Sol has an ECI of 161.8, Laguna XS.2 has not been scored yet and Claude Opus 5 has an ECI of 162.9.
Which is better for coding?
There are no published SWE-bench Verified results for GPT-5.6 Sol, Laguna XS.2 and Claude Opus 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-5.6 Sol has the largest context window at 1,050,000 tokens, against 1,000,000 for Claude Opus 5 and 262,144 for Laguna XS.2. Maximum output per response: GPT-5.6 Sol up to 128,000, Laguna XS.2 up to 32,768, Claude Opus 5 up to 128,000 tokens.
Which can read images, PDFs, audio or video?
GPT-5.6 Sol accepts text, images and PDFs; Laguna XS.2 accepts text; Claude Opus 5 accepts text, images and PDFs. GPT-5.6 Sol handles the widest range of inputs.
Are any of these open source?
Laguna XS.2 publishes its weights and can be self-hosted; GPT-5.6 Sol and Claude Opus 5 is proprietary.
Which is newer?
Claude Opus 5 is the newest, released Jul 24, 2026. GPT-5.6 Sol came out Jul 9, 2026; Laguna XS.2 came out Apr 28, 2026. Knowledge cutoff: GPT-5.6 Sol Feb 16, 2026, Claude Opus 5 May 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.
