GPT-5.5
OpenAI released 2026-04-24 proprietary
GPT-5.6 Sol is both better and cheaper.
It scores +4.6 higher and costs 64% less ($4.00/M against $11.25/M) on this workload — and it does everything this model does.
Grok 4.6 is cheaper still (73% less) but drops 1.1M → 500K context.
The same model via batch is 50% cheaper ($5.63/M) — same weights, different latency.
Every price dimension
| Input | $5.00/M |
|---|---|
| Output | $30.00/M |
| Cached input | $0.500/M |
| Web search | $0.01/call |
| Batch discount | 50% |
Past 272,000 tokens the price changes. Input goes to $10.00/M (2×) and output to $45.00/M. The headline rate does not apply to a long-context workload.
Independent scores
| Intelligence | 56.3 |
|---|---|
| Coding | 74.9 |
| Agentic | 47.4 |
| LMArena Elo | 1470.9high |
Sources are listed separately rather than averaged. Across the 62 models both have scored they correlate at r = 0.806 — close agreement overall, but four models rank very differently between them, and a blended score would hide exactly those.
Best rankings by task
- svg #7of 98 1274
- agenticslides #7of 16 1150
- agenticslides(python-pptx) #7of 16 1155
- pptxslides #7of 14 1157
- gamedev #8of 144 1337
- agentichtmlslides #9of 16 1084
- agenticslides(html) #9of 16 1077
- asciiart #10of 83 1282
Capability
| Context window | 1.1M |
|---|---|
| Max output | 128K |
| Input modes | file, image, text |
| Tool use | yes |
| Reasoning | optional |
| Knowledge cutoff | 2025-12-01 |
| Open weights | no |
Provenance
| Price source | openrouter.ai |
|---|---|
| Fetched | 2026-08-24 |
| Quality data | verified |
| Cross-checked | vendor page |
GPT-5.5 is OpenAI’s frontier model designed for complex professional workloads, building on GPT-5.4 with stronger reasoning, higher reliability, and improved token efficiency on hard tasks. It features a 1M+ token...