OpenAI, released Oct 6, 2025
GPT-5 Proprice, context, benchmarks and release details
64% confidence 64 percent, Medium confidence, 3 of 7 expected sources in
45.8
SI Score
#117 of 142 ranked models
- Input, per 1M tokens
- $15.00models.devFirst-party hosted API; MIT models.dev transcription. Provider documentation: https://platform.openai.com/docs/models. Exact canonical endpoint; lowest short-context Standard USD token tier; cache/batch discounts excluded. Deprecated endpoints excluded; openai/gpt-5-pro
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Output, per 1M tokens
- $120.00models.devFirst-party hosted API; MIT models.dev transcription. Provider documentation: https://platform.openai.com/docs/models. Exact canonical endpoint; lowest short-context Standard USD token tier; cache/batch discounts excluded. Deprecated endpoints excluded; openai/gpt-5-pro
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Context window
- 400Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Max output
- 272Kmodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - Released
- Oct 6, 2025OpenAI API changelogPublished source fact
Retrieved Oct 9, 2026 · factual citation
Open source ↗
How the score breaks down
Coding (weight 40 percent) —
Math (weight 15 percent) 37.7
Preference (weight 15 percent) —
Reasoning (weight 30 percent) 42.1
Weights: reasoning 30%, math 15%, coding 40%, preference 15%. Results use fixed 0–100 scales before averaging, and thin evidence is pulled toward 50. Method si-v3-retained-evidence-2, computed Oct 9, 2026, 05:13 UTC.
Around it on the leaderboard
- 115 Claude Sonnet 4 46.6
- 116 Nemotron 3.5 Lightning 30B A3B 46.1
- 117 GPT-5 Pro 45.8
- 118 Llama-3.1-70B-Instruct 45.7
- 119 Gemini 2.5 Pro 45.2
Benchmark results
7 benchmarks, 7 resultsEach row shows the best published result. Where a model was tested at several settings, such as reasoning effort, open the row to see each one. Hover or tap a value for its source.
ARC-AGI-1 (public eval)reasoning 77%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] gpt-5-pro-2025-10-06Published Oct 6, 2026
Retrieved Oct 9, 2026 · factual citation
Open source ↗ 77.0
Open source ↗ 77.0
ARC-AGI-1 (semi-private)reasoning 70.2%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] gpt-5-pro-2025-10-06Published Oct 6, 2026
Retrieved Oct 9, 2026 · factual citation
Open source ↗ 70.2
Open source ↗ 70.2
ARC-AGI-2 (public eval)reasoning 13.3%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] gpt-5-pro-2025-10-06Published Oct 6, 2026
Retrieved Oct 9, 2026 · factual citation
Open source ↗ 13.3
Open source ↗ 13.3
ARC-AGI-2 (semi-private)reasoning 18.3%ARC PrizeARC Prize steward-published result; exact edition/subset; publication time is HTTP Last-Modified of the aggregate export, not the evaluation date. [variant] gpt-5-pro-2025-10-06Published Oct 6, 2026
Retrieved Oct 9, 2026 · factual citation
Open source ↗ 18.3
Open source ↗ 18.3
Humanity's Last Exam (Scale AI)reasoning 31.6%Humanity’s Last ExamPotential contamination warning: This model was evaluated after the public release of HLE, allowing model builder access to the prompts and solutions. [variant] Published Nov 6, 2025
Retrieved Oct 9, 2026 · factual citation
Open source ↗ 31.6
Open source ↗ 31.6
FrontierMath Tier 4 (v2)math 19.5%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 19.5
Open source ↗ 19.5
FrontierMath Tiers 1–3 (v2)math 55.8%Epoch AI BenchmarkingEpoch-owned evaluation mean score; scores only, no benchmark questions [variant] highPublished Jun 12, 2026
Retrieved Oct 9, 2026 · CC-BY
Open source ↗ 55.8
Open source ↗ 55.8
Normalization uses a fixed 0–100 scale for each unit, independent of other models. Compare evaluation conditions before reading a small gap as decisive. “Lab-reported” marks the provider's own published figure.
Details and sources
- Open weights
- Nomodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - License
- not yet reported
- Input modalities
- text, imagemodels.devPublished source fact
Retrieved Oct 9, 2026 · MIT
Open source ↗ - First seen by SuperIndex
- Oct 8, 2026
- Coverage
- 51% of expected source weight
Reported (3)
- ARC PrizeOct 8, 2026
- Epoch AI BenchmarkingOct 8, 2026
- Humanity’s Last ExamOct 8, 2026
Awaiting (4)
- Official model cards via models.dev4% of weight
- LiveBench13% of weight
- LMArena / Arena25% of weight
- Terminal-Bench6% of weight
Confidence rises as pending sources publish. Some sources never cover some models, so confidence reaches 100% at 80% of expected weight.