Gemini 4 Argon vs GPT-5-Codex
Full profiles: Gemini 4 Argon (Google) and GPT-5-Codex (OpenAI): every benchmark, price and rank.
Updated · Data from 8 sources
As of 29 Oct 2026, Gemini 4 Argon is ahead of GPT-5-Codex on quality (65.9 vs 58.6 on the BenchLeader Index), but GPT-5-Codex is 2.3× cheaper at $3.44 vs $8.00 per 1M tokens blended. Choose Gemini 4 Argon for coding (67 vs 59) and long context (65 vs 61); choose GPT-5-Codex for fast first answers (10.9 s vs 47.3 s) and agents & tools (67 vs 59), where it also streams ~2× faster. Gemini 4 Argon's current price runs until 30 Nov 2026, then $8.00.
Key facts (as of 29 Oct 2026)
- Gemini 4 Argon scores 65.9 on the BenchLeader Index and GPT-5-Codex scores 58.6 (BenchLeader, 29 Oct 2026).
- Gemini 4 Argon costs $8.00 per 1M tokens blended until 30 Nov 2026, then $8.00 (https://ai.google.dev/gemini-api/docs/pricing, checked 20 Oct 2026).
- GPT-5-Codex costs $3.44 per 1M tokens blended (list price, 29 Oct 2026).
- GPT-5-Codex streams ~2× faster than Gemini 4 Argon (Artificial Analysis, 7-day median, 29 Oct 2026).
- Of the 4 benchmarks both models have, Gemini 4 Argon leads 4 (BenchLeader, 29 Oct 2026).
Gemini 4 Argon benchmarks 
GPT-5-Codex benchmarks Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Gemini 4 Argon | 67 vs 59 category score | High |
| Agents & tool use | GPT-5-Codex | 67 vs 59 category score | High |
| Long documents | Gemini 4 Argon | 65 vs 61 category score; AA-LCR v1.1 79.7% vs 71.7% | Medium |
| Chat quality (human preference) | – | Only Gemini 4 Argon has an Arena rating (72 category score) | Not comparable |
| Lowest cost | GPT-5-Codex | $3.44 vs $8.00 per 1M blended; agentic coding session $0.115 vs $0.320; from 1 Dec 2026: $3.44 vs $8.00 | – |
| Fastest first answer | GPT-5-Codex | 10.9 s (GPT-5-Codex (high)) vs 47.3 s (Gemini 4 Argon (high)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 4 Argon (high) leads on quality: 65.9 vs 58.6.
- Gemini 4 Argon (high) is stronger in coding, human preference, knowledge, long context, maths, reasoning.
- GPT-5-Codex (high) is stronger in agents & tools, instruction following, multimodal.
- GPT-5-Codex (high) is 2.3× cheaper ($3.44 vs $8.00 per 1M blended) until 30 Nov 2026, then $3.44 vs $8.00 from 1 Dec 2026.
- GPT-5-Codex (high) streams 2.4× faster (195 vs 83 tokens per second).
At a glance
| Spec | Gemini 4 Argon | GPT-5-Codex |
|---|---|---|
| Developer | OpenAI | |
| Released | 30 Sept 2026 | 15 Sept 2025 |
| Generally available | 13 Oct 2026 | 15 Sept 2025 |
| Context window (tokens) | 1,048,576 | 400,000 |
| Max output (tokens) | 65,536 (up to 1,048,576 with long decode) | – |
| Input types | text, image, video, audio, PDF | – |
| Reasoning effort settings | low, medium, high | – |
| Compared at | Gemini 4 Argon (high) | GPT-5-Codex (high) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $4.00 / $20.00 / $0.100 | $1.25 / $10.00 / $0.125 |
| Announced price change | $4.00 / $20.00 / $0.200 from 1 Dec 2026 | None announced |
| API model ID | gemini-4-argon | – |
| Where to use it | Gemini API (AI Studio), Gemini Enterprise Agent Platform, Gemini app | – |
Limits, availability and announced prices from the vendors' own pages: ai.google.dev, ai.google.dev pricing (checked 20 Oct 2026). Everything else is from the day's data.
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Gemini 4 Argon (high) | GPT-5-Codex (high) | Leader |
|---|---|---|---|
| Reasoning | 71.2 | – | not comparable |
| Coding | 66.5 | 58.5 | Gemini 4 Argon |
| Agents & tools | 58.8 | 66.8 | GPT-5-Codex |
| Maths | 70.3 | – | not comparable |
| Knowledge | 73.1 | – | not comparable |
| Instruction following | – | 72.4 | not comparable |
| Human preference | 71.6 | – | not comparable |
| Multimodal | – | 56.4 | not comparable |
| Long context | 64.7 | 60.5 | Gemini 4 Argon |
Where sources disagree
Across the 4 benchmarks both have, Gemini 4 Argon leads 4, GPT-5-Codex leads 0. Every overall index that rates both puts Gemini 4 Argon ahead.
| Overall index | Gemini 4 Argon (high) | GPT-5-Codex (high) | Leader |
|---|---|---|---|
| BenchLeader Index | 65.9 | 58.6 | Gemini 4 Argon |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Gemini 4 Argon (high) | GPT-5-Codex (high) |
|---|---|---|
| Blended price per 1M tokens, until 30 Nov 2026 | $8.00 | $3.44 |
| Input / output per 1M | $4.00 / $20.00 | $1.25 / $10.00 |
| Blended price per 1M from 1 Dec 2026 | $8.00 | $3.44 |
| Input / output per 1M from 1 Dec 2026 | $4.00 / $20.00 | $1.25 / $10.00 |
| Cached input per 1M | $0.100 | $0.125 |
| Price per task: chat reply | $0.0076 | $0.0035 |
| Price per task: summarize a 30-page report | $0.060 | $0.021 |
| Price per task: code edit | $0.054 | $0.022 |
| Price per task: agentic coding session | $0.320 | $0.115 |
| Price per task: structured extraction | $0.012 | $0.0045 |
| Cost to run a full benchmark suite | $7.68 | – |
| Output speed | 83 tok/s | 195 tok/s |
| Time to first answer | 47.3 s | 10.9 s |
| Time to a complete answer | 53.3 s | 13.4 s |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 4 Argon (high) scores 65.9 and takes 47 s to its first answer token; Gemini 4 Argon scores 63.1 in 5 s. GPT-5-Codex (high) scores 58.6 and takes 11 s to its first answer token. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
125 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Gemini 4 Argon (high) | GPT-5-Codex (high) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 65.9 | 58.6 |
| Agents & tools score | BenchLeader | 58.8 | 66.8 |
| Coding score | BenchLeader | 66.5 | 58.5 |
| Human preference score | BenchLeader | 71.6 | – |
| Knowledge score | BenchLeader | 73.1 | – |
| Long context score | BenchLeader | 64.7 | 60.5 |
| Maths score | BenchLeader | 70.3 | – |
| Reasoning score | BenchLeader | 71.2 | – |
| Instruction following score | BenchLeader | – | 72.4 |
| Multimodal score | BenchLeader | – | 56.4 |
| Blended price $/M | BenchLeader | $8.00 | $3.44 |
| Output speed | BenchLeader | 83 tok/s | 195 tok/s |
| Time to first answer | BenchLeader | 47.3 s | 10.9 s |
| Context window | BenchLeader | 1,048,576 tokens | 400,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 96.4% | – |
| Arena Hard Prompts | Arena (LMArena) | 1540 | – |
| LiveBench 2.0 Reasoning | LiveBench | 95.1%† | – |
| GPQA Diamond (AA) | Artificial Analysis | – | 83.7% |
| Humanity's Last Exam (AA) | Artificial Analysis | 57.1% | 27.9% |
| HLE-Diamond (AA) | Artificial Analysis | 56.7% | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | – | 70.3%† |
| ARC-AGI-1 | ARC Prize | 98.5% | – |
| ARC-AGI-2 | ARC Prize | 91.3% | – |
| ARC-AGI-3 | ARC Prize | 97.3% | – |
| CritPt | Artificial Analysis | 27.1% | 5.1% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 52.5% | – |
| MysteryMechanism | Vals AI | 45.5% | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 59.0% | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 64.0% | – |
| LMCA | Epoch AI Benchmarking Hub | 63.7% | – |
| DTBench | Epoch AI Benchmarking Hub | 97.3% | – |
| Coding | |||
| SciCode | SciCode | 61.8% | – |
| WeirdML | WeirdML | – | 54.5% |
| FrontierCode | Cognition | 47.5%† | – |
| Arena Coding | Arena (LMArena) | 1552 | – |
| Arena WebDev | Arena (LMArena) | 1680 | – |
| LiveBench 2.0 Coding | LiveBench | 89.7%† | – |
| LiveCodeBench | Vals AI | – | 84.7% |
| IOI | Vals AI | 100.0% | – |
Show all 125 benchmarks
| Metric | Publisher | Gemini 4 Argon (high) | GPT-5-Codex (high) |
|---|---|---|---|
| Coding | |||
| Code Migration | Vals AI | 68.2% | – |
| Code Migration v1.1 | Vals AI | 70.8% | – |
| ProgramBench | Vals AI | 2.5% | – |
| SWE-bench Multimodal | Vals AI | 56.0% | – |
| Vibe Code Bench 1-100 | Vals AI | 21.2% | – |
| Vibe Code Bench v1.1 | Vals AI | 91.9% | – |
| MirrorCode | Epoch AI Benchmarking Hub | 58.9% | – |
| CursorBench | Cursor | 52.9% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 2100.1 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | 55.0% | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 57.1% | – |
| Startrise Frontend | Startrise | 71.0%† | – |
| FrontierCode 1.1 (AA) | Artificial Analysis | 43.5% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 25.7% | – |
| LiveCodeBench (AA) | Artificial Analysis | – | 84.0% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | – | 44.3%† |
| Remote Labor Index | Scale AI / CAIS | 17.1%† | – |
| APEX-Agents | Mercor | 82.4%† | – |
| Arena Agent | Arena (LMArena) | 11.7 | – |
| LiveBench Long Agentic | LiveBench | 61.2%† | – |
| LiveBench Browser | LiveBench | 82.8%† | – |
| Terminal-Bench Hard | Artificial Analysis | – | 37.9% |
| τ²-Bench Telecom (AA) | Artificial Analysis | – | 86.8% |
| Terminal-Bench 2.1 (Vals) | Vals AI | 57.7% | – |
| HiL-Bench | Scale AI SEAL | 62.0%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 56.2% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1631 | – |
| Analyst Agent (AA) | Artificial Analysis | 48.8% | – |
| CUA-bench | Vals AI | 4.8% | – |
| CyberBench | Vals AI | 77.9% | – |
| Finance Agent v2 | Vals AI | 65.4% | – |
| Finance Agent v2.1 | Vals AI | 67.7% | – |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 27.1% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 19.6% | – |
| Legal Research Bench | Vals AI | 54.8% | – |
| Robo-bench | Vals AI | 0.8% | – |
| SREBench | Vals AI | 44.3% | – |
| Tax Agent Bench | Vals AI | 44.6% | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | 57.6% | – |
| Terminal-Bench Science | Vals AI | 44.3% | – |
| Time Horizon Index: KSP | Vals AI | 52.5% | – |
| Vending-Bench 2 | Andon Labs | 13718.2† | – |
| AutomationBench | Artificial Analysis | 77.5% | – |
| GDP.pdf (AA) | Artificial Analysis | 21.8% | – |
| Harvey LAB-AA v1.1 | Harvey | 3.9% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1483 | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 52.9% | – |
| FACTS Search | 84.0%† | – | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 86.0% | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 78.0% | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 100.0% | – |
| ProofBench | Vals AI | 99.0%† | – |
| Arena Maths | Arena (LMArena) | 1532 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 76.0% | – |
| LiveBench 2.0 Data Analysis | LiveBench | 85.2%† | – |
| LiveBench Research | LiveBench | 83.1%† | – |
| AA-Omniscience v1.1 | Artificial Analysis | 46.1 | – |
| LegalBench | Vals AI | 88.3% | – |
| PRBench Finance | Scale AI SEAL | 49.9%† | – |
| PRBench Legal | Scale AI SEAL | 49.4%† | – |
| BioMysteryBench | Vals AI | 76.3% | – |
| Excel Modeling Benchmark | Vals AI | 75.2% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 74.8% | – |
| MedCode | Vals AI | 58.8% | – |
| MedScribe | Vals AI | 87.4% | – |
| Public Benefits Bench v1.1 | Vals AI | 69.8% | – |
| DrugDiscoveryBench | Scale AI SEAL | 65.4% | – |
| EBR-bench | Epoch AI Benchmarking Hub | 64.8% | – |
| GDP.pdf | Epoch AI Benchmarking Hub | 28.4% | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 48.3% | – |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 95.7% | – |
| HealthBench Professional (AA) | OpenAI | 55.0% | – |
| Instruction following | |||
| IFBench | Artificial Analysis | – | 74.2% |
| MultiChallenge | Scale AI SEAL | 78.7%† | – |
| Arena Instruction Following | Arena (LMArena) | 1519 | – |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1517 | – |
| EQ-Bench 4 | EQ-Bench | 1236† | – |
| Arena Creative Writing | Arena (LMArena) | 1512 | – |
| Arena Multi-turn | Arena (LMArena) | 1544 | – |
| Arena Longer Queries | Arena (LMArena) | 1535 | – |
| Multimodal | |||
| LiveBench Design | LiveBench | 73.7%† | – |
| MMMU-Pro | Artificial Analysis | – | 73.8% |
| SAGE | Vals AI | 89.5% | – |
| AudioMC | Scale AI SEAL | 58.4% | – |
| AudioMC (text output) | Scale AI SEAL | 58.4% | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 54.5%† | – |
| Code2Video-AA | Artificial Analysis | 1259 | – |
| AA-3D | Artificial Analysis | 1153 | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 79.7% | 71.7% |
| MLCR | Artificial Analysis | 13.9% | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 166.4† | – |
| LiveBench 2.0 | LiveBench | 79.3%† | – |
| AA Intelligence Index v5.0 | Artificial Analysis | 50.4 | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 52.6 | 24.9 |
| AA Progress Index v1 | Artificial Analysis | 341.8 | 275.1 |
| Vals Index | Vals AI | 69.2 | – |
| Vals Finance Index | Vals AI | 64.9% | – |
| Vals RSI Index | Vals AI | 30.6% | – |
| Vals Science Index | Vals AI | 55.4% | – |
| Vals Router Index | Vals AI | 69.2%† | – |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- Gemini 4 Argon: list price moves to $4.00 / $20.00 per 1M input / output tokens (announced).
- Gemini 4 Argon (high): input price moved from $2.00/M to $4.00/M (100% up); output price moved from $10.00/M to $20.00/M (100% up); cost per task moved from $3.84/M to $7.68/M (100% up).
- Gemini 4 Argon (high): BenchLeader Index 66.8 → 65.9.
- Gemini 4 Argon (high): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 5 score revisions; cost per task moved from $1.62/M to $3.84/M (137% up).
- GPT-5-Codex (high): 1 new result (AA Progress Index v1).
- Gemini 4 Argon (high): BenchLeader Index 68.5 → 66.8.
- Gemini 4 Argon (high): 1 new result (Terminal-Bench 2.1 (Vals)).
- Gemini 4 Argon (high): cost per task moved from $1.99/M to $1.62/M (19% down).
- Gemini 4 Argon (high): 10 new results (Code Migration v1.1, Excel Modeling Benchmark v1.1, Finance Agent v2.1…); 7 score revisions.
- Gemini 4 Argon (high): BenchLeader Index 69.0 → 68.5.
Related comparisons
- Gemini 4 Argon vs Claude Opus 5.5
- Gemini 4 Argon vs GPT-6 Astra
- Gemini 4 Argon vs GPT-6.1 Sol
- GPT-5-Codex vs Claude Opus 5.5
- GPT-5-Codex vs GPT-6 Astra
- GPT-5-Codex vs GPT-6.1 Sol
- GPT-5-Codex vs Claude Fable 5.1
- Gemini 4 Argon vs GPT-5 Pro
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 4 Argon vs GPT-5-Codex: questions
- Is Gemini 4 Argon better than GPT-5-Codex?
- Gemini 4 Argon (high) leads on quality: 65.9 vs 58.6. The BenchLeader Index combines every independent quality benchmark; Gemini 4 Argon (high) is ahead overall as of 29 Oct 2026, but check the category scores for your use.
- Gemini 4 Argon vs GPT-5-Codex: which is cheaper, and will that change?
- GPT-5-Codex is cheaper today: $3.44 against $8.00 per million tokens, blended at three input tokens per output token. Gemini 4 Argon's current price runs until 30 Nov 2026, then $8.00.
- Is Gemini 4 Argon or GPT-5-Codex better for coding?
- Gemini 4 Argon scores higher in coding (67 vs 59 on the category index, where 50 is average).
- Which is faster, Gemini 4 Argon or GPT-5-Codex?
- GPT-5-Codex streams ~2× faster (195 vs 83 output tokens per second, Artificial Analysis 7-day median).
- Is Gemini 4 Argon or GPT-5-Codex better for agents and tool use?
- GPT-5-Codex scores higher in agentic tasks (67 vs 59 on the category index, where 50 is average).
More questions (4)
- Which is better for long documents?
- Gemini 4 Argon scores higher on long-context benchmarks (65 vs 61). Gemini 4 Argon accepts more context: 1,048,576 against 400,000 tokens for GPT-5-Codex.
- Which has lower latency?
- GPT-5-Codex answers first: 10.9 s against 47.3 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Gemini 4 Argon vs GPT-5-Codex?
- Across the 4 benchmarks both have, Gemini 4 Argon leads 4, GPT-5-Codex leads 0. Every overall index that rates both puts Gemini 4 Argon ahead.
- When was Gemini 4 Argon released?
- Gemini 4 Argon became generally available on 13 Oct 2026, after first appearing on 30 Sept 2026.