← Model list
GPT-5.5
openai·openai/gpt-5.5·BETA·Closed·gpt series
Default frontier GPT for coding, computer use, research, and knowledge work
At a glance
Official price$5.00 / $30.00 per 1M
Cache read $0.50 · OpenAI
Lowest paid$0.188 / $1.13
UnoRouter Gateway · 33.3× spread
2 more $0 channels
Context1,050,000
⚠ Providers report 272,000–1,050,000; the table below is authoritative
Output limit128,000
Capabilities
✓ Reasoning✓ Tool use✓ Structured outputTemperature✓ Attachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Knowledge cutoff2025-12-01
Released / updated2026-04-23 / 2026-04-23
Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier xhigh
Intelligence56.3
Value133
Coding74.9
Agentic47.4
Output speed85 tok/s
TTFT68.86 s
Cost per task$1.1747
Value formulaIQ 56.3 ÷ min blended $0.422 = 133
Reasoning tier → intelligence / speed (higher tier = stronger but slower)
Non-reasoningIQ 35.8 · 68 tok/s
lowIQ 44.5 · 74 tok/s
mediumIQ 51.4 · 79 tok/s
highIQ 54.7 · 74 tok/s
xhighIQ 56.3 · 85 tok/s
Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.
Available at 45 providers41 with public prices · 2 free
| Provider | Tier | Input | Output | Cache read | Cache write | Context | Output limit | Status |
|---|---|---|---|---|---|---|---|---|
| Kenari gpt-5-5 | Gateway | Free | — | — | 1,050,000 | 128,000 | ||
| UnoRouter gpt-5.5:free | Gateway | Free | — | — | 1,050,000 | 128,000 | ||
| UnoRouter | Gateway | $0.188 | $1.13 | — | — | 1,050,000 | 128,000 | |
| Xpersona | Gateway | $1.50 | $12.00 | $0.15 | — | 1,050,000 | 128,000 | |
| FrogBot gpt-5-5 | Gateway | $2.50 | $15.00 | $0.25 | — | 272,000 ⚠ | 128,000 | |
| Xpersona xpersona-gpt-5.5 | Gateway | $3.00 | $18.00 | $0.30 | — | 1,000,000 ⚠ | 128,000 | |
| Poe | Gateway | $4.55 | $27.27 | $0.455 | — | 400,000 ⚠ | 128,000 | |
| Requesty | Gateway | $4.95 | $29.70 | $0.495 | — | 1,050,000 | 128,000 | |
| HPC-AI | Cloud | $5.00 tiered >272K: $10.00 | $30.00 | $0.50 | — | 1,050,000 | 128,000 | |
| AI-ROUTER | Gateway | $5.00 | $30.00 | $0.50 | — | 1,050,000 | 128,000 | |
| OpenAIOfficial | First-party | $5.00 tiered >272K: $10.00 | $30.00 | $0.50 | — | 1,050,000 | 128,000 | |
Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.
Your usage cost
1UnoRouter$93.75
2Xpersona$738.00
3FrogBot$980.00
4Xpersona$1,176.00
5Poe$1,781.82
6Requesty$1,940.40
16OpenAI · Official$1,960.00
2 more channels offer $0 (Kenari, UnoRouter); free tiers usually have rate limits and no SLA, excluded from ranking.
Switch to UnoRouter to save $1,866.25/mo (95%)
Note: this is a gateway; verify availability and rate limits yourself.
Benchmark31 items
| Name | Conditions | Score | Metric | Source |
|---|---|---|---|---|
| SWE-Bench Pro | — | 58.6 | resolve rate | Source ↗ |
| Terminal-Bench | harness: Terminus-2 · v2.1 | 78.2 | success rate | Source ↗ |
| SWE-Atlas Codebase QnA | harness: Codex | 45.43 | score | Source ↗ |
| SWE-Atlas Refactoring | harness: Codex | 44.79 | score | Source ↗ |
| SWE-Atlas Test Writing | harness: Codex | 42.59 | score | Source ↗ |
| Artificial Analysis Coding Agent Index | harness: Codex · variant: xhigh | 65.3 | average pass@1 | Source ↗ |
| SWE-Atlas Codebase QnA | harness: Codex · variant: xhigh | 80.8 | pass@1 | Source ↗ |
| SWE-Bench Pro | harness: Codex · variant: xhigh · dataset: hard-aa | 30.9 | pass@1 | Source ↗ |
| Terminal-Bench | harness: Codex · variant: xhigh · v2.1 | 84.1 | pass@1 | Source ↗ |
| Artificial Analysis Coding Agent Index | harness: Codex · variant: medium | 60.4 | average pass@1 | Source ↗ |
| SWE-Atlas Codebase QnA | harness: Codex · variant: medium | 79.1 | pass@1 | Source ↗ |
| SWE-Bench Pro | harness: Codex · variant: medium · dataset: hard-aa | 26.2 | pass@1 | Source ↗ |
| Terminal-Bench | harness: Codex · variant: medium · v2.1 | 75.8 | pass@1 | Source ↗ |
| Artificial Analysis Coding Agent Index | harness: Cursor CLI · variant: medium | 57.8 | average pass@1 | Source ↗ |
| SWE-Atlas Codebase QnA | harness: Cursor CLI · variant: medium | 75 | pass@1 | Source ↗ |
| SWE-Bench Pro | harness: Cursor CLI · variant: medium · dataset: hard-aa | 24.9 | pass@1 | Source ↗ |
| Terminal-Bench | harness: Cursor CLI · variant: medium · v2.1 | 73.4 | pass@1 | Source ↗ |
| Terminal-Bench | v2.0 | 82.7 | success rate | Source ↗ |
| GPQA Diamond | — | 93.6 | accuracy | Source ↗ |
| Humanity's Last Exam | variant: no tools | 41.4 | accuracy | Source ↗ |
| Humanity's Last Exam | variant: with tools | 52.2 | accuracy | Source ↗ |
| OSWorld-Verified | — | 78.7 | success rate | Source ↗ |
| BrowseComp | — | 84.4 | accuracy | Source ↗ |
| MMMU Pro | variant: no tools | 81.2 | accuracy | Source ↗ |
| ARC-AGI-2 | variant: Verified | 85 | accuracy | Source ↗ |
| FrontierMath | dataset: Tier 1-3 | 51.7 | accuracy | Source ↗ |
| FrontierMath | dataset: Tier 4 | 35.4 | accuracy | Source ↗ |
| GDPval | — | 84.9 | wins or ties | Source ↗ |
| MCP Atlas | — | 75.3 | success rate | Source ↗ |
| Toolathlon | — | 55.6 | success rate | Source ↗ |
| τ²-Bench Telecom | variant: original prompts | 98 | success rate | Source ↗ |
The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.
Reasoning control
effort = noneeffort = loweffort = mediumeffort = higheffort = xhigheffort = maxbudget_tokenseffort = minimal
4 / 45 providers expose no reasoning control (reasoning_options: []).
Experimental modes6 items
| Mode | Provider | Input | Output | Cache read | Cache write |
|---|---|---|---|---|---|
| fast | OrcaRouter | $12.50 | $75.00 | $1.25 | — |
| fast | Databricks | $12.50 | $75.00 | $1.25 | — |
| fast | OpenAI | $12.50 | $75.00 | $1.25 | — |
| fast | Merge Gateway | $12.50 | $75.00 | $1.25 | — |
| fast | ZenMux | $12.50 | $75.00 | $1.25 | — |
| fast | AIHubMix | $12.50 | $75.00 | $1.25 | — |
Related models
RouteLLMsame series$3.00 / $15.00GPT-Realtime-2.1same series$4.00 / $24.00NanoGPT Helpsame series$0 / $0GLM-5.2cheaper alternative$1.40 / $4.40DeepSeek V4 Procheaper alternative$0.435 / $0.87DeepSeek V4 Flashcheaper alternative$0.14 / $0.28Kimi K3cheaper alternative$3.00 / $15.00
Price historyone sample accumulated per data sync
Input listOutput listMin blended