OpenAIAuto-detectedGPT-4 Turbo
Best configuration ranks #496 of 610 on the BenchLeader Index at 41.0 ±4.8. Last measured 2 Sept 2026. Released 9 Apr 2024.
- Blended price
- $15.00/M
- $10.00 in · $30.00 out
- Output speed
- 31 tok/s
- First answer
- 3.14 s
- first token 1.58 s
- Context
- 128k
- Overall index41
- Reasoning36
- Coding38
- Maths32
- Human preference49
- Multimodal39
- Composite38
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| high | – | – | 31 tok/s | 3.14 s | $0.013 | Knowledge 51 |
| defaultbest | 41.0 | #496 | 31 tok/s | 3.14 s | $0.013 | Coding 38 · Composite 38 · Human preference 49 · Maths 32 · Multimodal 39 · Reasoning 36 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | high | default | Source |
|---|---|---|---|
| GPQA Diamond | – | 46.6%#214 | Epoch AI Benchmarking Hub |
| SimpleBench | – | 25.1%#72 | SimpleBench |
| LMArena Hard Prompts | – | 1316#226 | LMArena |
| Humanity's Last Exam (AA)not in index | – | 3.1%#513 | Artificial Analysis |
Coding
| Benchmark | high | default | Source |
|---|---|---|---|
| WeirdML | – | 18.0%#138 | WeirdML |
| LMArena Coding | – | 1347#224 | LMArena |
Maths
| Benchmark | high | default | Source |
|---|---|---|---|
| FrontierMath Tiers 1–3 | – | 0.7%#100 | Epoch AI Benchmarking Hub |
| OTIS Mock AIME | – | 6.7%#220 | Epoch AI Benchmarking Hub |
| MATH Level 5 | – | 46.7%#55 | Epoch AI Benchmarking Hub |
Knowledge
| Benchmark | high | default | Source |
|---|---|---|---|
| LegalBench | 80.5%#78 | – | Vals AI |
| MedQA | 82.0%#66 | – | Vals AI |
Human preference
| Benchmark | high | default | Source |
|---|---|---|---|
| LMArena Text | – | 1324#206 | LMArena |
Multimodal
| Benchmark | high | default | Source |
|---|---|---|---|
| LMArena Vision | – | 1090#109 | LMArena |
Composite
| Benchmark | high | default | Source |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | – | 127.3#143 | Epoch AI Benchmarking Hub |
| AA Intelligence Index | – | 7.0#416 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| OpenAI | 27 tok/s | 1.58 s | $10.00 | $30.00 | 128k | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.013 | – | 12.8 s |
| Summarise a 30-page report | 12,000 / 600 | $0.138 | – | 22.4 s |
| Code edit | 6,000 / 1,500 | $0.105 | – | 51.3 s |
| Agentic coding session | 60,000 / 4,000 | $0.720 | – | 2.2 min |
| Structured extraction | 2,000 / 200 | $0.026 | – | 9.6 s |
See also
Data as of 9 Sept 2026. Compare these configurations.