● gittrendingglobal

Home › AI model benchmarks

Best AI models by use case, explained simply

Updated 2026-10-17 18:00 IST · By the Git Trending Global team · Every number links its source

Quick answer

There is no single best AI model. On the tests below, the leaders are: coding: Claude Opus 5.5 on LMArena WebDev; reasoning and math: Gemini 4 Argon on LiveBench 2.0 Reasoning; chat and writing: Gemini 4 Argon on LMArena Text; speed and price: Gemini 3.5 Flash-Lite on Output speed. Pick the use case you care about and tap a model to see its price, size and every score with its source.

Best for coding

Top on LMArena WebDev: Claude Opus 5.5 (1807 Arena score).

LMArena WebDev

When people ask two AIs to build the same website or app and compare the results, which one do they pick?

Higher is better. Tap a model to see everything we know about it.

  1. 1Claude Opus 5.5 independent1807 Arena score

    Claude Opus 5.5 (Anthropic): 1807 Arena score independent · setting: Max

    as of 2026-10-17 · source

    All details for Claude Opus 5.5

  2. 2GPT-6 Astra independent1777 Arena score

    GPT-6 Astra (OpenAI): 1777 Arena score independent · setting: Max

    as of 2026-10-17 · source

    All details for GPT-6 Astra

  3. 3Claude Fable 5.1 independent1741 Arena score

    Claude Fable 5.1 (Anthropic): 1741 Arena score independent · setting: Max

    as of 2026-10-17 · source

    All details for Claude Fable 5.1

  4. 4Gemini 4 Argon independent1678 Arena score

    Gemini 4 Argon (Google): 1678 Arena score independent · setting: High

    as of 2026-10-17 · source

    All details for Gemini 4 Argon

  5. 5Grok 4.7 independent1639 Arena score

    Grok 4.7 (xAI): 1639 Arena score independent · setting: xHigh

    as of 2026-10-17 · source

    All details for Grok 4.7

  6. 6DeepSeek-V4.1-Flash independent1618 Arena score

    DeepSeek-V4.1-Flash (DeepSeek): 1618 Arena score independent · setting: Max

    as of 2026-10-17 · source

    All details for DeepSeek-V4.1-Flash

  7. 7Qwen3.8-27B independent1593 Arena score

    Qwen3.8-27B (Alibaba (Qwen)): 1593 Arena score independent

    as of 2026-10-17 · source

    All details for Qwen3.8-27B

  8. 8Claude Haiku 5.5 independent1585 Arena score

    Claude Haiku 5.5 (Anthropic): 1585 Arena score independent · setting: High

    as of 2026-10-17 · source

    All details for Claude Haiku 5.5

  9. 9GPT-6 Luna independent1578 Arena score

    GPT-6 Luna (OpenAI): 1578 Arena score independent · setting: Max

    as of 2026-10-17 · source

    All details for GPT-6 Luna

  10. 10Gemini 3.5 Flash-Lite independent1441 Arena score

    Gemini 3.5 Flash-Lite (Google): 1441 Arena score independent

    as of 2026-10-17 · source

    All details for Gemini 3.5 Flash-Lite

  11. 11Gemma 4 31B independent1367 Arena score

    Gemma 4 31B (Google): 1367 Arena score independent

    as of 2026-10-17 · source

    All details for Gemma 4 31B

  12. 12Grok 4.3 independent1358 Arena score

    Grok 4.3 (xAI): 1358 Arena score independent

    as of 2026-10-17 · source

    All details for Grok 4.3

  13. 13Mistral Medium 3.5 independent1264 Arena score

    Mistral Medium 3.5 (Mistral AI): 1264 Arena score independent

    as of 2026-10-17 · source

    All details for Mistral Medium 3.5

Source: LMArena (arena.ai), as of 2026-10-17 · independent test

Default view with style control. Scores within each model's plus or minus range are close to tied. Each model's best listed setting is used.

LiveBench 2.0 Coding

Fresh coding tasks with tests that check the answer, changed often so models cannot memorize them.

Higher is better. Tap a model to see everything we know about it.

  1. 1Claude Fable 5.1 independent97.7 score out of 100

    Claude Fable 5.1 (Anthropic): 97.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Fable 5.1

  2. 2Claude Opus 5.5 independent96.3 score out of 100

    Claude Opus 5.5 (Anthropic): 96.3 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Opus 5.5

  3. 3GPT-6 Astra independent93.7 score out of 100

    GPT-6 Astra (OpenAI): 93.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for GPT-6 Astra

  4. 4Gemini 4 Argon independent89.7 score out of 100

    Gemini 4 Argon (Google): 89.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Gemini 4 Argon

  5. 5Qwen3.8-27B independent86.8 score out of 100

    Qwen3.8-27B (Alibaba (Qwen)): 86.8 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Qwen3.8-27B

  6. 6DeepSeek-V4.1-Flash independent85.4 score out of 100

    DeepSeek-V4.1-Flash (DeepSeek): 85.4 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for DeepSeek-V4.1-Flash

  7. 7GPT-6 Luna independent84.5 score out of 100

    GPT-6 Luna (OpenAI): 84.5 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for GPT-6 Luna

  8. 8Claude Haiku 5.5 independent84.5 score out of 100

    Claude Haiku 5.5 (Anthropic): 84.5 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Haiku 5.5

  9. 9Grok 4.7 independent84.5 score out of 100

    Grok 4.7 (xAI): 84.5 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Grok 4.7

  10. 10Gemma 4 31B independent49.0 score out of 100

    Gemma 4 31B (Google): 49.0 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Gemma 4 31B

Source: LiveBench 2.0, as of 2026-10-14 · independent test

LiveBench 2.0 release of 2 Sep 2026, shown on the site on 14 Oct 2026. Models run at their highest supported effort.

Best for reasoning and math

Top on LiveBench 2.0 Reasoning: Gemini 4 Argon (95.1 score out of 100).

LiveBench 2.0 Reasoning

Fresh logic and puzzle questions with one right answer, changed often so models cannot memorize them.

Higher is better. Tap a model to see everything we know about it.

  1. 1Gemini 4 Argon independent95.1 score out of 100

    Gemini 4 Argon (Google): 95.1 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Gemini 4 Argon

  2. 2GPT-6 Astra independent92.7 score out of 100

    GPT-6 Astra (OpenAI): 92.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for GPT-6 Astra

  3. 3Claude Opus 5.5 independent92.2 score out of 100

    Claude Opus 5.5 (Anthropic): 92.2 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Opus 5.5

  4. 4Claude Fable 5.1 independent91.7 score out of 100

    Claude Fable 5.1 (Anthropic): 91.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Fable 5.1

  5. 5DeepSeek-V4.1-Flash independent86.7 score out of 100

    DeepSeek-V4.1-Flash (DeepSeek): 86.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for DeepSeek-V4.1-Flash

  6. 6Claude Haiku 5.5 independent84.6 score out of 100

    Claude Haiku 5.5 (Anthropic): 84.6 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Claude Haiku 5.5

  7. 7Grok 4.7 independent82.7 score out of 100

    Grok 4.7 (xAI): 82.7 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Grok 4.7

  8. 8GPT-6 Luna independent81.8 score out of 100

    GPT-6 Luna (OpenAI): 81.8 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for GPT-6 Luna

  9. 9Qwen3.8-27B independent80.0 score out of 100

    Qwen3.8-27B (Alibaba (Qwen)): 80.0 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Qwen3.8-27B

  10. 10Gemma 4 31B independent68.3 score out of 100

    Gemma 4 31B (Google): 68.3 score out of 100 independent · setting: highest supported effort

    as of 2026-10-14 · source

    All details for Gemma 4 31B

Source: LiveBench 2.0, as of 2026-10-14 · independent test

LiveBench 2.0 release of 2 Sep 2026, shown on the site on 14 Oct 2026. Models run at their highest supported effort.

Artificial Analysis Intelligence Index

One combined score from many independent tests of how smart a model is.

Higher is better. Tap a model to see everything we know about it.

  1. 1Claude Opus 5.5 independent58 index points

    Claude Opus 5.5 (Anthropic): 58 index points independent · setting: max with fallback

    as of 2026-10-17 · source

    All details for Claude Opus 5.5

  2. 2GPT-6 Astra independent53 index points

    GPT-6 Astra (OpenAI): 53 index points independent · setting: max

    as of 2026-10-17 · source

    All details for GPT-6 Astra

  3. 3Claude Fable 5.1 independent53 index points

    Claude Fable 5.1 (Anthropic): 53 index points independent · setting: max with fallback

    as of 2026-10-17 · source

    All details for Claude Fable 5.1

  4. 4Gemini 4 Argon independent53 index points

    Gemini 4 Argon (Google): 53 index points independent · setting: high

    as of 2026-10-17 · source

    All details for Gemini 4 Argon

  5. 5Grok 4.7 independent46 index points

    Grok 4.7 (xAI): 46 index points independent · setting: xhigh

    as of 2026-10-17 · source

    All details for Grok 4.7

  6. 6Claude Haiku 5.5 independent43 index points

    Claude Haiku 5.5 (Anthropic): 43 index points independent · setting: max

    as of 2026-10-17 · source

    All details for Claude Haiku 5.5

  7. 7DeepSeek-V4.1-Flash independent39 index points

    DeepSeek-V4.1-Flash (DeepSeek): 39 index points independent · setting: max

    as of 2026-10-17 · source

    All details for DeepSeek-V4.1-Flash

  8. 8GPT-6 Luna independent38 index points

    GPT-6 Luna (OpenAI): 38 index points independent · setting: max

    as of 2026-10-17 · source

    All details for GPT-6 Luna

  9. 9Qwen3.8-27B independent34 index points

    Qwen3.8-27B (Alibaba (Qwen)): 34 index points independent · setting: xhigh

    as of 2026-10-17 · source

    All details for Qwen3.8-27B

  10. 10Gemini 3.5 Flash-Lite independent22 index points

    Gemini 3.5 Flash-Lite (Google): 22 index points independent

    as of 2026-10-17 · source

    All details for Gemini 3.5 Flash-Lite

  11. 11Gemma 4 31B independent15 index points

    Gemma 4 31B (Google): 15 index points independent

    as of 2026-10-17 · source

    All details for Gemma 4 31B

  12. 12Mistral Medium 3.5 independent14 index points

    Mistral Medium 3.5 (Mistral AI): 14 index points independent

    as of 2026-10-17 · source

    All details for Mistral Medium 3.5

Source: Artificial Analysis, as of 2026-10-17 · independent test

Each model's highest-scoring setting is used. The page shows no index version or date; as_of is the day it was read.

GPQA Diamond

Very hard science questions in biology, chemistry and physics, written by PhD experts.

Higher is better. Tap a model to see everything we know about it.

  1. 1GPT-6 Astra maker's own number96.0%

    GPT-6 Astra (OpenAI): 96.0% maker's own number

    as of 2026-09-03 · source

    All details for GPT-6 Astra

  2. 2Claude Opus 5.5 maker's own number91%

    Claude Opus 5.5 (Anthropic): 91% maker's own number

    as of 2026-10-17 · source

    All details for Claude Opus 5.5

  3. 3DeepSeek-V4.1-Flash maker's own number90.9%

    DeepSeek-V4.1-Flash (DeepSeek): 90.9% maker's own number

    as of 2026-10-17 · source

    All details for DeepSeek-V4.1-Flash

  4. 4Qwen3.8-27B maker's own number89.2%

    Qwen3.8-27B (Alibaba (Qwen)): 89.2% maker's own number

    as of 2026-10-17 · source

    All details for Qwen3.8-27B

  5. 5Claude Haiku 5.5 maker's own number85%

    Claude Haiku 5.5 (Anthropic): 85% maker's own number

    as of 2026-10-17 · source

    All details for Claude Haiku 5.5

  6. 6Gemma 4 31B maker's own number84.3%

    Gemma 4 31B (Google): 84.3% maker's own number

    as of 2026-10-17 · source

    All details for Gemma 4 31B

Source: Makers' own model cards, as of 2026-10-17 · numbers can include makers' own claims

Maker's own numbers, run with different settings, so compare with care. Close to saturated at the top and questioned for possible training leaks. Anthropic's figures are whole numbers from its cost guide.

Humanity's Last Exam (no tools)

2,500 expert questions written to be too hard for AI, answered without web search or code.

Higher is better. Tap a model to see everything we know about it.

  1. 1Claude Opus 5.5 maker's own number64.4%

    Claude Opus 5.5 (Anthropic): 64.4% maker's own number

    as of 2026-09-22 · source

    All details for Claude Opus 5.5

  2. 2Claude Fable 5.1 maker's own number60.9%

    Claude Fable 5.1 (Anthropic): 60.9% maker's own number

    as of 2026-09-22 · source

    All details for Claude Fable 5.1

  3. 3Claude Haiku 5.5 maker's own number45.9%

    Claude Haiku 5.5 (Anthropic): 45.9% maker's own number

    as of 2026-10-07 · source

    All details for Claude Haiku 5.5

  4. 4DeepSeek-V4.1-Flash maker's own number36.8%

    DeepSeek-V4.1-Flash (DeepSeek): 36.8% maker's own number

    as of 2026-10-17 · source

    All details for DeepSeek-V4.1-Flash

  5. 5Qwen3.8-27B maker's own number30.8%

    Qwen3.8-27B (Alibaba (Qwen)): 30.8% maker's own number

    as of 2026-10-17 · source

    All details for Qwen3.8-27B

  6. 6Gemma 4 31B maker's own number19.5%

    Gemma 4 31B (Google): 19.5% maker's own number

    as of 2026-10-17 · source

    All details for Gemma 4 31B

Source: Makers' own model cards, as of 2026-10-17 · numbers can include makers' own claims

The official leaderboard lists none of these models yet, so these are the makers' own numbers, graded by different judges. Some answers in the test are disputed.

Best for chat and writing

Top on LMArena Text: Gemini 4 Argon (1524 Arena score).

LMArena Text

When real people compare two AI answers side by side without knowing which AI wrote them, which one do they pick more often?

Higher is better. Tap a model to see everything we know about it.

  1. 1Gemini 4 Argon independent1524 Arena score

    Gemini 4 Argon (Google): 1524 Arena score independent · setting: High

    as of 2026-10-15 · source

    All details for Gemini 4 Argon

  2. 2Claude Opus 5.5 independent1508 Arena score

    Claude Opus 5.5 (Anthropic): 1508 Arena score independent · setting: High

    as of 2026-10-15 · source

    All details for Claude Opus 5.5

  3. 3Claude Fable 5.1 independent1502 Arena score

    Claude Fable 5.1 (Anthropic): 1502 Arena score independent · setting: Max

    as of 2026-10-15 · source

    All details for Claude Fable 5.1

  4. 4GPT-6 Astra independent1477 Arena score

    GPT-6 Astra (OpenAI): 1477 Arena score independent · setting: Max

    as of 2026-10-15 · source

    All details for GPT-6 Astra

  5. 5DeepSeek-V4.1-Flash independent1475 Arena score

    DeepSeek-V4.1-Flash (DeepSeek): 1475 Arena score independent · setting: Max

    as of 2026-10-15 · source

    All details for DeepSeek-V4.1-Flash

  6. 6Gemini 3.5 Flash-Lite independent1455 Arena score

    Gemini 3.5 Flash-Lite (Google): 1455 Arena score independent

    as of 2026-10-15 · source

    All details for Gemini 3.5 Flash-Lite

  7. 7Gemma 4 31B independent1452 Arena score

    Gemma 4 31B (Google): 1452 Arena score independent

    as of 2026-10-15 · source

    All details for Gemma 4 31B

  8. 8Grok 4.7 independent1444 Arena score

    Grok 4.7 (xAI): 1444 Arena score independent · setting: xHigh

    as of 2026-10-15 · source

    All details for Grok 4.7

  9. 9GPT-6 Luna independent1443 Arena score

    GPT-6 Luna (OpenAI): 1443 Arena score independent · setting: Max

    as of 2026-10-15 · source

    All details for GPT-6 Luna

  10. 10Grok 4.3 independent1442 Arena score

    Grok 4.3 (xAI): 1442 Arena score independent

    as of 2026-10-15 · source

    All details for Grok 4.3

  11. 11Qwen3.8-27B independent1438 Arena score

    Qwen3.8-27B (Alibaba (Qwen)): 1438 Arena score independent

    as of 2026-10-15 · source

    All details for Qwen3.8-27B

  12. 12Mistral Medium 3.5 independent1426 Arena score

    Mistral Medium 3.5 (Mistral AI): 1426 Arena score independent

    as of 2026-10-15 · source

    All details for Mistral Medium 3.5

Source: LMArena (arena.ai), as of 2026-10-15 · independent test

Default view with style control. Each score has a plus or minus range, so models within about 10 points are close to tied. Each model's best listed setting is used.

Best for speed and price

Top on Output speed: Gemini 3.5 Flash-Lite (449 tokens per second).

Output speed

How many tokens (pieces of words) per second the model writes once it starts answering. Higher is faster.

Higher is better. Tap a model to see everything we know about it.

  1. 1Gemini 3.5 Flash-Lite independent449 tokens per second

    Gemini 3.5 Flash-Lite (Google): 449 tokens per second independent

    as of 2026-10-17 · source

    All details for Gemini 3.5 Flash-Lite

  2. 2Claude Haiku 5.5 independent235 tokens per second

    Claude Haiku 5.5 (Anthropic): 235 tokens per second independent · setting: max

    as of 2026-10-17 · source

    All details for Claude Haiku 5.5

  3. 3DeepSeek-V4.1-Flash independent176 tokens per second

    DeepSeek-V4.1-Flash (DeepSeek): 176 tokens per second independent · setting: max

    as of 2026-10-17 · source

    All details for DeepSeek-V4.1-Flash

  4. 4Mistral Medium 3.5 independent160 tokens per second

    Mistral Medium 3.5 (Mistral AI): 160 tokens per second independent

    as of 2026-10-17 · source

    All details for Mistral Medium 3.5

  5. 5GPT-6 Luna independent120 tokens per second

    GPT-6 Luna (OpenAI): 120 tokens per second independent · setting: max

    as of 2026-10-17 · source

    All details for GPT-6 Luna

  6. 6Claude Opus 5.5 independent97 tokens per second

    Claude Opus 5.5 (Anthropic): 97 tokens per second independent · setting: max with fallback

    as of 2026-10-17 · source

    All details for Claude Opus 5.5

  7. 7Grok 4.7 independent93 tokens per second

    Grok 4.7 (xAI): 93 tokens per second independent · setting: xhigh

    as of 2026-10-17 · source

    All details for Grok 4.7

  8. 8Claude Fable 5.1 independent71 tokens per second

    Claude Fable 5.1 (Anthropic): 71 tokens per second independent · setting: max with fallback

    as of 2026-10-17 · source

    All details for Claude Fable 5.1

  9. 9GPT-6 Astra independent57 tokens per second

    GPT-6 Astra (OpenAI): 57 tokens per second independent · setting: max

    as of 2026-10-17 · source

    All details for GPT-6 Astra

  10. 10Qwen3.8-27B independent41 tokens per second

    Qwen3.8-27B (Alibaba (Qwen)): 41 tokens per second independent · setting: xhigh

    as of 2026-10-17 · source

    All details for Qwen3.8-27B

  11. 11Gemma 4 31B independent35 tokens per second

    Gemma 4 31B (Google): 35 tokens per second independent

    as of 2026-10-17 · source

    All details for Gemma 4 31B

Source: Artificial Analysis, as of 2026-10-17 · independent test

Median speed for the same setting used in the Intelligence Index. Gemini 4 Argon has no speed listed yet.

Compare all models

Scroll sideways on a phone. Empty cells mean we have no number from a source, not a zero. Score columns show tests with 3 or more models.

ModelCompanyPrice in / out (per million tokens)Context (tokens)Open weightsLMArena WebDevLiveBench 2.0 CodingLiveBench 2.0 ReasoningArtificial Analysis Intelligence IndexGPQA DiamondHumanity's Last Exam (no tools)LMArena TextOutput speed
GPT-6 AstraOpenAI$10 / $501,050,000No1777 Arena score93.7 score out of 10092.7 score out of 10053 index points96.0%1477 Arena score57 tokens per second
GPT-6 LunaOpenAI$0.1 / $0.51,050,000No1578 Arena score84.5 score out of 10081.8 score out of 10038 index points1443 Arena score120 tokens per second
Claude Opus 5.5Anthropic$4 / $20No1807 Arena score96.3 score out of 10092.2 score out of 10058 index points91%64.4%1508 Arena score97 tokens per second
Claude Fable 5.1Anthropic$10 / $501,000,000No1741 Arena score97.7 score out of 10091.7 score out of 10053 index points60.9%1502 Arena score71 tokens per second
Claude Haiku 5.5Anthropic$0.1 / $0.51,000,000No1585 Arena score84.5 score out of 10084.6 score out of 10043 index points85%45.9%235 tokens per second
Gemini 4 ArgonGoogle$2 / $101,048,576No1678 Arena score89.7 score out of 10095.1 score out of 10053 index points1524 Arena score
Gemini 3.5 Flash-LiteGoogle$0.3 / $2.51,048,576No1441 Arena score22 index points1455 Arena score449 tokens per second
Grok 4.7xAI$2 / $6500,000No1639 Arena score84.5 score out of 10082.7 score out of 10046 index points1444 Arena score93 tokens per second
Grok 4.3xAI$1.25 / $2.51,000,000No1358 Arena score1442 Arena score
DeepSeek-V4.1-FlashDeepSeek$0.3 / $1.21,000,000Yes1618 Arena score85.4 score out of 10086.7 score out of 10039 index points90.9%36.8%1475 Arena score176 tokens per second
Qwen3.8-27BAlibaba (Qwen)262,144Yes1593 Arena score86.8 score out of 10080.0 score out of 10034 index points89.2%30.8%1438 Arena score41 tokens per second
Gemma 4 31BGoogle262,144Yes1367 Arena score49.0 score out of 10068.3 score out of 10015 index points84.3%19.5%1452 Arena score35 tokens per second
Mistral Medium 3.5Mistral AI262,144Yes1264 Arena score14 index points1426 Arena score160 tokens per second

Every model, in detail

GPT-6 Astra · OpenAI

Near the top for coding and hard reasoning; slower to start on its deepest settings.

Company
OpenAI
Released
2026-09-03
Open weights
No
Context
1,050,000 tokens
Price per million tokens
$10 in / $50 out · price source

All its scores

  • LMArena Text: 1477 Arena score (Max) independentas of 2026-10-15 · source
  • LMArena WebDev: 1777 Arena score (Max) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 53 index points (max) independentas of 2026-10-17 · source
  • Output speed: 57 tokens per second (max) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 92.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 93.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 96.0% maker's own numberas of 2026-09-03 · source

Official page for GPT-6 Astra

GPT-6 Luna · OpenAI

Very cheap and quick for simple everyday jobs.

Company
OpenAI
Released
2026-09-22
Open weights
No
Context
1,050,000 tokens
Price per million tokens
$0.1 in / $0.5 out · price source

All its scores

  • LMArena Text: 1443 Arena score (Max) independentas of 2026-10-15 · source
  • LMArena WebDev: 1578 Arena score (Max) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 38 index points (max) independentas of 2026-10-17 · source
  • Output speed: 120 tokens per second (max) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 81.8 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 84.5 score out of 100 (highest supported effort) independentas of 2026-10-14 · source

Official page for GPT-6 Luna

Claude Opus 5.5 · Anthropic

Top pick for building apps and websites, and the highest overall intelligence score here.

Company
Anthropic
Released
2026-09-22
Open weights
No
Price per million tokens
$4 in / $20 out · price source

All its scores

  • LMArena Text: 1508 Arena score (High) independentas of 2026-10-15 · source
  • LMArena WebDev: 1807 Arena score (Max) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 58 index points (max with fallback) independentas of 2026-10-17 · source
  • Output speed: 97 tokens per second (max with fallback) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 92.2 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 96.3 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 91% maker's own numberas of 2026-10-17 · source
  • Humanity's Last Exam (no tools): 64.4% maker's own numberas of 2026-09-22 · source

Official page for Claude Opus 5.5

Claude Fable 5.1 · Anthropic

Strong all-rounder for writing and coding, but one of the priciest.

Company
Anthropic
Released
2026-09-01
Open weights
No
Context
1,000,000 tokens
Price per million tokens
$10 in / $50 out · price source

All its scores

  • LMArena Text: 1502 Arena score (Max) independentas of 2026-10-15 · source
  • LMArena WebDev: 1741 Arena score (Max) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 53 index points (max with fallback) independentas of 2026-10-17 · source
  • Output speed: 71 tokens per second (max with fallback) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 91.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 97.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • Humanity's Last Exam (no tools): 60.9% maker's own numberas of 2026-09-22 · source

Official page for Claude Fable 5.1

Claude Haiku 5.5 · Anthropic

Fast and very cheap, with surprisingly strong reasoning for its price.

Company
Anthropic
Released
2026-10-07
Open weights
No
Context
1,000,000 tokens
Price per million tokens
$0.1 in / $0.5 out · price source

All its scores

  • LMArena WebDev: 1585 Arena score (High) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 43 index points (max) independentas of 2026-10-17 · source
  • Output speed: 235 tokens per second (max) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 84.6 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 84.5 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 85% maker's own numberas of 2026-10-17 · source
  • Humanity's Last Exam (no tools): 45.9% maker's own numberas of 2026-10-07 · source

Official page for Claude Haiku 5.5

Gemini 4 Argon · Google

Most preferred by people for everyday chat and writing, and the best fresh-puzzle reasoning score.

Company
Google
Released
2026-10-13
Open weights
No
Context
1,048,576 tokens
Price per million tokens
$2 in / $10 out · price source

All its scores

  • LMArena Text: 1524 Arena score (High) independentas of 2026-10-15 · source
  • LMArena WebDev: 1678 Arena score (High) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 53 index points (high) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 95.1 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 89.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source

Official page for Gemini 4 Argon

Gemini 3.5 Flash-Lite · Google

The fastest model here, for quick and simple tasks.

Company
Google
Released
2026-07-21
Open weights
No
Context
1,048,576 tokens
Price per million tokens
$0.3 in / $2.5 out · price source

All its scores

  • LMArena Text: 1455 Arena score independentas of 2026-10-15 · source
  • LMArena WebDev: 1441 Arena score independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 22 index points independentas of 2026-10-17 · source
  • Output speed: 449 tokens per second independentas of 2026-10-17 · source

Official page for Gemini 3.5 Flash-Lite

Grok 4.7 · xAI

Solid mid-range model at a moderate price.

Company
xAI
Released
2026-09-21
Open weights
No
Context
500,000 tokens
Price per million tokens
$2 in / $6 out · price source

All its scores

  • LMArena Text: 1444 Arena score (xHigh) independentas of 2026-10-15 · source
  • LMArena WebDev: 1639 Arena score (xHigh) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 46 index points (xhigh) independentas of 2026-10-17 · source
  • Output speed: 93 tokens per second (xhigh) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 82.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 84.5 score out of 100 (highest supported effort) independentas of 2026-10-14 · source

Official page for Grok 4.7

Grok 4.3 · xAI

An older, cheaper xAI model; it scores below the newer ones.

Company
xAI
Open weights
No
Context
1,000,000 tokens
Price per million tokens
$1.25 in / $2.5 out · price source

All its scores

  • LMArena Text: 1442 Arena score independentas of 2026-10-15 · source
  • LMArena WebDev: 1358 Arena score independentas of 2026-10-17 · source

Official page for Grok 4.3

DeepSeek-V4.1-Flash · DeepSeek

The strongest open model here: free to download, cheap to use and quick.

Company
DeepSeek
Released
2026-09-10
Open weights
Yes, you can download it
Context
1,000,000 tokens
Price per million tokens
$0.3 in / $1.2 out · price source

All its scores

  • LMArena Text: 1475 Arena score (Max) independentas of 2026-10-15 · source
  • LMArena WebDev: 1618 Arena score (Max) independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 39 index points (max) independentas of 2026-10-17 · source
  • Output speed: 176 tokens per second (max) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 86.7 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 85.4 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 90.9% maker's own numberas of 2026-10-17 · source
  • Humanity's Last Exam (no tools): 36.8% maker's own numberas of 2026-10-17 · source

Official page for DeepSeek-V4.1-Flash

Qwen3.8-27B · Alibaba (Qwen)

A capable open model small enough to run on a good home computer.

Company
Alibaba (Qwen)
Released
2026-08-14
Open weights
Yes, you can download it
Context
262,144 tokens

All its scores

  • LMArena Text: 1438 Arena score independentas of 2026-10-15 · source
  • LMArena WebDev: 1593 Arena score independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 34 index points (xhigh) independentas of 2026-10-17 · source
  • Output speed: 41 tokens per second (xhigh) independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 80.0 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 86.8 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 89.2% maker's own numberas of 2026-10-17 · source
  • Humanity's Last Exam (no tools): 30.8% maker's own numberas of 2026-10-17 · source

Official page for Qwen3.8-27B

Gemma 4 31B · Google

Google's open model you can run yourself; best for simple tasks.

Company
Google
Released
2026-03-31
Open weights
Yes, you can download it
Context
262,144 tokens

All its scores

  • LMArena Text: 1452 Arena score independentas of 2026-10-15 · source
  • LMArena WebDev: 1367 Arena score independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 15 index points independentas of 2026-10-17 · source
  • Output speed: 35 tokens per second independentas of 2026-10-17 · source
  • LiveBench 2.0 Reasoning: 68.3 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • LiveBench 2.0 Coding: 49.0 score out of 100 (highest supported effort) independentas of 2026-10-14 · source
  • GPQA Diamond: 84.3% maker's own numberas of 2026-10-17 · source
  • Humanity's Last Exam (no tools): 19.5% maker's own numberas of 2026-10-17 · source

Official page for Gemma 4 31B

Mistral Medium 3.5 · Mistral AI

Open model that answers quickly, but scores lower on hard tests.

Company
Mistral AI
Released
2026-04-28
Open weights
Yes, you can download it
Context
262,144 tokens

All its scores

  • LMArena Text: 1426 Arena score independentas of 2026-10-15 · source
  • LMArena WebDev: 1264 Arena score independentas of 2026-10-17 · source
  • Artificial Analysis Intelligence Index: 14 index points independentas of 2026-10-17 · source
  • Output speed: 160 tokens per second independentas of 2026-10-17 · source

Official page for Mistral Medium 3.5

What we left out on purpose

Sources

Updated 2026-10-17 18:00 IST. Scores and prices come only from these pages and each model's own source links above:

📲 Want the news when a new model tops these charts? Follow @gittrendingglobal on Instagram, or read today's AI news.