Artificial Analysis
The most recommended of 10 AI benchmarks
Runs standardized tests on AI models and combines results into comparable intelligence, cost and speed charts. Critics say its weighting and biases shape rankings.
Runs standardized tests on AI models and combines results into comparable intelligence, cost and speed charts. Critics say its weighting and biases shape rankings.
108 people gave an opinion
- recommend it
- 62
- criticize it
- 46
- approval share
- 57%
Recommended in 11 of 13 months
- November 2025: not recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: not recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: recommended (month in progress)
Why it doesn’t carry the badge
- 62 people recommend itNeeds 10 or more, met.
- 1.3 recommenders for every criticNeeds 2 to 1 or better, not met.
- Recommended in 11 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Cost and speed versus intelligence charts
“I recommend looking at the "Intelligence vs. Cost to Run Artificial Analysis Intelligence Index"”
Comment from 2 May 2026 in “DeepSeek V4–almost on the frontier, a fraction of the price” “benchmarks that include speed across many models and providers available at various places”
Comment from 17 Jun 2026 in “Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?” “Personally I do trust them, but I mainly look at the Intelligence Index vs. Time per Task”
Comment from 23 Oct 2026 in “Kimi K3.1 Pro and K3.1 Pro Highspeed” - Comment from 3 Oct 2026 in “One month coding with GLM 5.3 Flash”
- Comment from 16 Jul 2026 in “Kimi K3 is now live”
- Comment from 24 Sep 2026 in “Best LLM for every budget, updated daily”
- Comment from 8 Aug 2026 in “Qwen3.8 Max now ranked as the best overall model by agentic index”
Many benchmarks combined in one place
“The most popular with broad coverage of a range of benchmarks is artificialanalysis.ai”
Comment from 19 Oct 2026 in “MiniMax M3.1 Flash” “AA seems to do a good job making a single number as representative as possible”
Comment from 26 Aug 2026 in “GLM-5.3-Flash” “They are the only site that really combines all benchmarks into a single place”
Comment from 23 Oct 2026 in “Kimi K3.1 Pro and K3.1 Pro Highspeed” - Comment from 25 Jul 2026 in “Opus 5 is currently #1 on Artificial Analysis Intelligence Leaderboard”
- Comment from 22 Apr 2026 in “Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model”
Current frontier models and prices
“Look up price versus intelligence for your favorite open model on artificial analysis”
Comment from 1 Nov 2026 in “ChatGPT and Codex usage limits merging” “Artificial Analysis is usually a good resource to find current frontier AI models.”
Comment from 10 Oct 2026 in “Triple-A Minesweeper” “check out LLM service prices on https://artificialanalysis.ai/”
Comment from 8 Dec 2025 in “The "confident idiot" problem: Why AI needs hard rules, not vibe checks” - Comment from 24 Apr 2026 in “DeepSeek v4”
Independent testing instead of lab claims
“consistently running standardized benchmarks for LLMs, rather than just aggregating unverified claims from the AI labs”
Comment from 11 Feb 2026 in “GLM-OCR: Accurate × Fast × Comprehensive” “a solid and independent, though also to be fair somewhat imperfect source for a general overview”
Comment from 29 Jan 2026 in “OpenAI's Unit Economics” “this is the whole raison d'etre for independent testing, like https://artificialanalysis.ai”
Comment from 9 Jul 2026 in “Introducing Muse Spark 1.1”
What people warn about
Each mark is one comment
Inconsistent model inclusion and timing
“Artificialanalysis should have a policy of only including available models or something.”
Comment from 28 Oct 2026 in “Gemini 4 Carbon” “some models are on the leaderboard second they're released while others come days later”
Comment from 19 Oct 2026 in “DeepSeek V4.1 Flash 1017 (Max)” “Also ArtificialAnalysis drop older foundational models to make room for new ones.”
Comment from 5 Sep 2026 in “Artificial Analysis Intelligence Index v4.2” - Comment from 22 Sep 2026 in “Claude Opus 5.5 Intelligence, Performance and Price Analysis (Max)”
Biased or opaque methodology
“Artificial Analysis has some very specific biases or perspectives on what they are measuring, so I wouldn't take their benchmarks as the final word on model quality.”
Comment from 9 Oct 2026 in “Step 5 Preview, a 1M-context MoE from StepFun, shows up on OpenRouter” “I don't trust these Artificial Analysis indexes and I'm not sure why anyone considers them authoritative”
Comment from 23 Oct 2026 in “Kimi K3.1 Pro and K3.1 Pro Highspeed” “treat AA results as just a data point and not the overarching 'authority' they may intend to be.”
Comment from 30 Oct 2026 in “Artificial Analysis v5 and the benchmarker's dilemma” - Comment from 4 Sep 2026 in “GPT-6 Astra”
Missing higher reasoning effort levels
“the reasoning efforts are available for the Intelligence Index, but not the Coding Agents benchmark I was interested in”
Comment from 30 Oct 2026 in “Artificial Analysis v5 and the benchmarker's dilemma” “it's really unfortunate that artificial analysis only provides comparison at the highest reasoning level”
Comment from 15 Oct 2026 in “Muse Spark Open” “I wish AA benchmarked the models at the reasoning efforts people actual use them at.”
Comment from 30 Oct 2026 in “Artificial Analysis v5 and the benchmarker's dilemma” Chart and ranking display errors
“it's irresponsible for them to put the points on the chart if they don't have the correct values for one of the axes yet”
Comment from 8 Oct 2026 in “Claude Haiku 5.5” “Why does the top card say "Intelligence #1/173" when the bar chart further down shows it only at position 7?”
Comment from 26 Aug 2026 in “GLM-5.3-Flash Intelligence, Performance and Price Analysis” “they lower the bars of tied scores so on the site it looks like Astra in second place”
Comment from 10 Sep 2026 in “Cognition launches new SWE-2 model, Rivaling Fable 5.1 and GPT-Astra”
Other AI benchmarks people recommend
The four most recommended other AI benchmarks, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| Artificial AnalysisThis page | 62 | 46 | 57% | 11 | — | — |
| LMArena | 39 | 40 | 49% | 11 | — | — |
| ARC-AGI | 22 | 24 | 48% | 8 | — | — |
| OCR Arena | 11 | 4 | 73% | 4 | — | Approved |
| Bullshit Benchmark | 9 | 0 | 100% | 6 | — | — |
What people said
40 of the 119 comments that mention it, newest first. Each links to the original comment.
“Look up price versus intelligence for your favorite open model on artificial analysis”
“treat AA results as just a data point and not the overarching 'authority' they may intend to be.”
“the reasoning efforts are available for the Intelligence Index, but not the Coding Agents benchmark I was interested in”
“AA is the best website I found with benchmarks numbers on all reasoning effort levels of recent frontier models.”
“I wish AA benchmarked the models at the reasoning efforts people actual use them at.”
“Artificialanalysis should have a policy of only including available models or something.”
“Personally I do trust them, but I mainly look at the Intelligence Index vs. Time per Task”
“Its a benchmark collection. When a benchmark gets saturated, they remove it from the index and add new ones.”
“They are the only site that really combines all benchmarks into a single place”
“I don't trust these Artificial Analysis indexes and I'm not sure why anyone considers them authoritative”
“The most popular with broad coverage of a range of benchmarks is artificialanalysis.ai”
“some models are on the leaderboard second they're released while others come days later”
“it's really unfortunate that artificial analysis only provides comparison at the highest reasoning level”
“I have mixed feelings about Artificial Analysis but they at least let you filter by size class and by proprietary versus open”
“Artificial Analysis is usually a good resource to find current frontier AI models.”
“Artificial Analysis has some very specific biases or perspectives on what they are measuring, so I wouldn't take their benchmarks as the final word on model quality.”
“it's irresponsible for them to put the points on the chart if they don't have the correct values for one of the axes yet”
“My two favorite graphs are AA's A Index vs Time Per Task and AA Index vs Cost Per Task.”
“No other model on Artificial Analysis dropped harder after they adjusted their weighting.”
“Artificial Analysis' own cost per task is a more fair estimation of cost.”
“ArtificialAnalysis tweaks stuff until newest big proprietary model is on top”
“they lower the bars of tied scores so on the site it looks like Astra in second place”
“Also ArtificialAnalysis drop older foundational models to make room for new ones.”
“I think it's in their interest to keep it confusing and opaque. Not exactly independent.”
“Why does the top card say "Intelligence #1/173" when the bar chart further down shows it only at position 7?”
“AA seems to do a good job making a single number as representative as possible”
“clearly shows both the intelligence and speed differences”
“there is literally all the metrics so you can make an informed decision”
“Theres value in some of AAs charts, like cost per job, and how often it hallucinated”
“I always look at things like Artificial Analysis' total cost to run their tests.”
“this is the whole raison d'etre for independent testing, like https://artificialanalysis.ai”
“benchmarks that include speed across many models and providers available at various places”
“I recommend looking at the "Intelligence vs. Cost to Run Artificial Analysis Intelligence Index"”
“The first result from Artificial Analysis shows good things”
“they merge multiple benchmarks across weighted categories”
“consistently running standardized benchmarks for LLMs, rather than just aggregating unverified claims from the AI labs”
“a solid and independent, though also to be fair somewhat imperfect source for a general overview”
“Artificial Analysis at least has the same at 20% from the top, so maybe that's the one we all agree to use for now since it implies faster growth.”
“Two others you can trust, off the top of my head, are LiveBench.ai and Artifical Analysis.”
“check out LLM service prices on https://artificialanalysis.ai/”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.