GPT-5 vs Llama 4 Scout
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Wins 1 of 7 areas
Facts
GPT-5 is the stronger all-rounder.Llama 4 Scout is cheaper and better at facts.
Scores updated · 32 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- CodingWriting and fixing software40GPT-54 of 4 tests
- ReasoningHard problems that need careful thinking40GPT-54 of 4 tests
- AgentsCarrying out multi-step tasks on its own20GPT-52 of 2 tests
- Images and chartsUnderstanding pictures, charts and video20GPT-52 of 2 tests
- Long documentsFinding answers in very long texts10GPT-51 of 1 test
- Following instructionsDoing exactly what it is asked10GPT-51 of 1 test
- FactsGetting facts right instead of making them up12Llama 4 Scout2 of 3 tests
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Llama 4 Scout costs 92% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where GPT-5 pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+65.8points ahead
- Reasons across sets of long documentsAA-LCR+50.5points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+33.6points ahead
Where Llama 4 Scout pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+2.8points ahead
Every test, side by side
All 32 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5
- SWE-bench VerifiedGPT-5 by 65.874.99.1+65.8
- Terminal-Bench 2.1GPT-5 by 31.535.23.7+31.5
- Terminal-Bench HardGPT-5 by 31.132.61.5+31.1
- SciCodeGPT-5 by 21.642.921.3+21.6
ReasoningGPT-5
- GPQA DiamondGPT-5 by 26.785.458.7+26.7
- Humanity's Last ExamGPT-5 by 24.728.53.8+24.7
- ARC-AGI-2GPT-5 by 9.99.90+9.9
- CritPtGPT-5 by 5.75.70+5.7
FactsLlama 4 Scout
- AA-Omniscience · AccuracyGPT-5 by 25.140.315.2+25.1
- Vectara HHEM hallucination ratelower is betterLlama 4 Scout by 7.415.17.7+7.4
- AA-Omniscience · Non-hallucinationLlama 4 Scout by 2.817.820.6+2.8
Images and chartsGPT-5
Other results15 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- τ²-Bench Telecom (AA run)GPT-5 by 69.384.815.5+69.3
- LiveCodeBenchGPT-5 by 50.783.532.8+50.7
- AA-OmniscienceGPT-5 by 43.4-8.7-52.1+43.4
- HarmBenchGPT-5 by 37.697.660+37.6
- AIR-Bench 2024GPT-5 by 35.487.752.3+35.4
- Artificial Analysis Coding IndexGPT-5 by 29.637.88.2+29.6
- AA IntelligenceGPT-5 by 14.9238.1+14.9
- MMLU-ProGPT-5 by 12.887.174.3+12.8
- bbqGPT-5 by 9.396.887.5+9.3
- vectara_factual_consistencyLlama 4 Scout by 7.484.992.3+7.4
- simple_safety_testsGPT-5 by 2.899.897+2.8
- anthropic_red_teamGPT-5 by 2.699.196.5+2.6
- vectara_avg_summary_lengthGPT-5 by 25.4 rating points162.7137.3+25.4 rating
- XSTestGPT-5 by 1.597.195.6+1.5
- vectara_answer_ratetie99.999tie
Questions people ask
Which is better, GPT-5 or Llama 4 Scout?
GPT-5 wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Llama 4 Scout wins facts, and costs 92% less.
Which is better for coding?
GPT-5. It wins 4 of the 4 coding tests both models report; Llama 4 Scout wins none.
Which is cheaper?
GPT-5 costs $1.25 per million input tokens and $10.00 per million output tokens; Llama 4 Scout costs $0.19 and $0.68. That makes Llama 4 Scout about 92% cheaper for the same work.
How do you compare the two?
We use the 32 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.