Grok 4.6 vs GPT-5.6 Sol
GPT-5.6 Sol comes out ahead on 10 of 14 shared measures, and the two rank 12 and 35 of 51 overall. Grok 4.6 is 2.7× cheaper per million tokens (blended) — A$8.14 vs A$19.65 per 1,000 replies of customer support replys.
Where each one wins.
GPT-5.6 Sol ranks 12th of 51; Grok 4.6 ranks 35th.
GPT-5.6 Sol ranks 7th of 42; Grok 4.6 ranks 9th.
Grok 4.6 ranks 4th of 41; GPT-5.6 Sol ranks 6th.
GPT-5.6 Sol ranks 7th of 51; Grok 4.6 ranks 10th.
GPT-5.6 Sol ranks 6th of 51; Grok 4.6 ranks 34th.
Grok 4.6 ranks 15th of 50; GPT-5.6 Sol ranks 22nd.
Grok 4.6 is 2.7× cheaper per million tokens (blended) — A$8.14 vs A$19.65 per 1,000 replies of customer support replys.
Grok 4.6 writes 1.1× faster (70 tok/s vs 62 tok/s).
Both can be used from Australia through their makers’ own APIs (xAI and OpenAI), which may process requests outside Australia. Grok 4.6 can be called from AWS Bedrock in Sydney and Melbourne, Microsoft Azure in Australia East and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. GPT-5.6 Sol can be called from AWS Bedrock in Sydney and Melbourne and Microsoft Azure in Australia East, but only through global routing, so requests may be processed outside Australia.
Side by side.
“Level” means the gap is inside LMArena’s confidence interval, or the scores are identical. Independent test scores are for each model’s headline setting. Both models can be used from Australia through their makers’ APIs; the Australian rows show whether requests can also be processed in Australia.
| BENCHMARK | GROK 4.6 | GPT-5.6 SOL | LEADS |
|---|---|---|---|
| AA Intelligence Index | 44.3 | 47.0 | GPT-5.6 Sol |
| AA Coding Index | 76.8 | 77.4 | GPT-5.6 Sol |
| GPQA Diamond | 94.9% | 94.1% | Grok 4.6 |
| Humanity’s Last Exam | 42.9% | 49.5% | GPT-5.6 Sol |
| AA-LCR | 80.3% | 84.0% | GPT-5.6 Sol |
| Terminal-Bench 2.1 | 88.4% | 88.0% | Grok 4.6 |
| τ²-Bench (banking) | 50.7% | 44.3% | Grok 4.6 |
| SciCode | 56.5% | 57.1% | GPT-5.6 Sol |
| LMArena Text | 1456 | 1484 | GPT-5.6 Sol |
| LMArena WebDev | 1616 | 1617 | Level |
| LMArena Vision | 1265 | 1286 | GPT-5.6 Sol |
| LMArena Document | 1461 | 1489 | GPT-5.6 Sol |
| LMArena Search | 1256 | ||
| LMArena Agent | #21 | #7 | GPT-5.6 Sol |
| Blended price (AUD) | A$4.21 | A$11.23 | Grok 4.6 |
| Output speed | 70 tok/s | 62 tok/s | Grok 4.6 |
| Answer starts after | 35 s | 1.0 min | Grok 4.6 |
| Context window | 500,000 tokens | 1,050,000 tokens | GPT-5.6 Sol |
| USE FROM AUSTRALIA | |||
| Maker’s API | YES | YES | |
| Processed in Australia | NO | NO | |
| AUSTRALIAN CLOUD REGIONS | |||
| AWS Bedrock · Sydney | GLOBAL | GLOBAL | |
| AWS Bedrock · Melbourne | GLOBAL | GLOBAL | |
| Azure · Australia East | GLOBAL | GLOBAL | |
| Google Vertex AI · Sydney | GLOBAL | NOT OFFERED | |
The same jobs, in Australian dollars.
List prices at the snapshot’s RBA rate, excluding GST and reasoning tokens.
| JOB | GROK 4.6 | GPT-5.6 SOL | MEDIAN MODEL |
|---|---|---|---|
| Customer support replyPER 1,000 REPLIES | A$8.14 | A$19.65 | ≈ A$6.58 |
| Summarise a 30-page documentPER 100 DOCUMENTS | A$6.29 | A$13.48 | ≈ A$4.74 |
| Agentic coding taskPER 10 TASKS | A$4.89 | A$10.67 | ≈ A$3.72 |
Each model in full.
Grok 4.6 or GPT-5.6 Sol?
Is Grok 4.6 better than GPT-5.6 Sol?
On our overall leaderboard GPT-5.6 Sol ranks 12th and Grok 4.6 35th of 51. Which is better depends on the job: the table above compares each benchmark, price and Australian availability.
Which is cheaper, Grok 4.6 or GPT-5.6 Sol?
Grok 4.6, at A$4.21 per million tokens (blended) against A$11.23 for GPT-5.6 Sol. Reasoning-heavy settings add output tokens, so real bills depend on the effort level you run.
Can I use Grok 4.6 and GPT-5.6 Sol in Australia?
Both can be used from Australia through their makers’ own APIs (xAI and OpenAI), which may process requests outside Australia. Grok 4.6 can be called from AWS Bedrock in Sydney and Melbourne, Microsoft Azure in Australia East and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. GPT-5.6 Sol can be called from AWS Bedrock in Sydney and Melbourne and Microsoft Azure in Australia East, but only through global routing, so requests may be processed outside Australia.
Ratings: LMArena leaderboard dataset (CC BY 4.0), rescaled for the lens scores · leaderboards to 22 Sep 2026
Evaluations, prices and speed: Artificial Analysis (artificialanalysis.ai) · fetched 23 Sep 2026
Exchange rate: Reserve Bank of Australia, table F11.1 · A$1 = US$0.7123 on 22 Sep 2026 · prices exclude GST
Snapshot 23 Sep 2026 · updated weekly · How the rankings work →