Updated Nov 11, 2026

AI Model Leaderboard, November 2026

Every major AI model ranked by quality, speed, pricing, and value. Filter by category, sort by any metric, and find the right model for your use case. Live data refreshed regularly with LMSys Arena Elo, official provider pricing, and Artificial Analysis benchmarks.

In short:As of November 2026, Claude Fable 5.2 leads the AI model leaderboard at 100/100 on our composite quality index, across 79+ models ranked by quality, price, speed, and value. The best pick depends on the workload: sort the table by the metric that matters to you.

🥇

Claude Fable 5.2

100/100

🥈

Claude Mythos 5.2

100/100

🥉

Claude Fable 5

99/100

Climb the Leaderboard

Stop reading: start ranking

Monthly Snapshot

November 2026: Top Models, Best Value, Fastest Inference

The November 2026 ranking covers 79 models across LMSys Arena Elo, MMLU Pro, HumanEval, MATH, pricing, and inference speed. Top of the table: Claude Fable 5.2 at 100/100 quality. The full table below is sortable by any metric. Live data is refreshed regularly from official provider pricing pages and the public Arena.

Top 5 by Quality Index

  1. Claude Fable 5.2: 100/100
  2. Claude Mythos 5.2: 100/100
  3. Claude Fable 5: 99/100
  4. Claude Opus 5: 99/100
  5. Claude Mythos 5: 99/100

Best Price-to-Quality

  1. DeepSeek V4 Flash: $0.28/1M out
  2. GLM-5.3 Flash: $0.5/1M out
  3. Gemini 2.0 Flash: $0.4/1M out
  4. Claude Haiku 5.5: $0.5/1M out
  5. GPT-6 Luna: $0.5/1M out

See our LMSys Arena deep dive and the monthly release roundup.

79 models
#ModelQualityArena ELOSpeedPriceContextValueReleased
1

Anthropic · Hardest long-horizon agentic work, cost no object

100
—101 t/s$8 / $401M4.2Nov 2026
2

Anthropic · Ceiling capability (verified organisations only)

100
——$8 / $401M4.2Nov 2026
3

Anthropic · Frontier agentic coding & knowledge work

99
152558 t/s$10 / $501M3.3Jun 2026
4

Anthropic · Frontier agentic coding & reasoning

99
152274 t/s$5 / $251M6.6Jul 2026
5

Anthropic · Ceiling capability (limited access)

99
153156 t/s$10 / $501M3.3Jun 2026
6

OpenAI · Frontier software engineering & computer use

99
—53 t/s$10 / $501M3.3Sep 2026
7

Anthropic · Top-tier reasoning with heavy prompt reuse

99
—68 t/s$10 / $501M3.3Sep 2026
8

Anthropic · Top-scoring frontier reasoning & agentic coding

99
1509—$4 / $201M8.3Sep 2026
9

Anthropic · High-scoring agents when you can cap effort

99
——$2 / $101M16.5Sep 2026
10

Google · Frontier reasoning at a sub-Fable price (intro rate to 30 Nov)

99
—78 t/s$2 / $101M16.5Nov 2026
11

Anthropic · Coding, agents & computer use

98
151272 t/s$5 / $251M6.5May 2026
12

OpenAI · General frontier reasoning & tools

98
151496 t/s$5 / $301M5.6Jul 2026
13

OpenAI · Near-Astra agentic quality at a fifth of the price

98
——$2 / $101M16.3Sep 2026
14

OpenAI · Frontier general purpose

97
150670 t/s$5 / $301M5.5Apr 2026
15

Moonshot AI · Open-weight coding frontier

97
150055 t/s$3 / $151M10.8Jul 2026
16

Meta · High-throughput batch inference

97
—255 t/s$1.25 / $4.251M35.3Sep 2026
17

OpenAI · Low-cost GPT-6 generation (see GPT-6.1 Sol)

97
——$2 / $101M16.2Sep 2026
18

Moonshot AI · Best open-weight agentic model if you can host 2.8T

97
—32 t/sSelf-host1M, Oct 2026
19

OpenAI · Reasoning at any cost

96
151068 t/s$30 / $1801M0.9Apr 2026
20

Anthropic · Coding & agentic workflows

96
150568 t/s$5 / $251M6.4Apr 2026
21

Google · Science & long-context

96
1505131 t/s$2 / $121M13.7Apr 2026
22

Alibaba Cloud · Multimodal APAC frontier (hosted); text-only open weights

96
149186 t/s$2 / $61M24.0Aug 2026
23

Z.ai (Zhipu AI) · Highest-scoring open-weight flagship

96
—67 t/s$1.4 / $4.41M33.1Aug 2026
24

Google · Long-horizon coding and migrations at a mid-tier price

96
—66 t/s$2 / $101M16.0Oct 2026
25

OpenAI · Computer-use and office agents at $1/$5

96
—74 t/s$1 / $51M32.0Oct 2026
26

xAI · Balanced reasoning under 500k context

95
—65 t/s$2 / $6500K23.8Aug 2026
27

OpenAI · Hard reasoning

94
137068 t/s$2 / $8200K18.8Apr 2025
28

Alibaba Cloud · Long autonomous agentic runs

94
148890 t/s$2.5 / $7.51M18.8May 2026
29

xAI · Cheap output tokens & real-time info

94
149988 t/s$2 / $6500K23.5Jul 2026
30

xAI · Agentic tasks & real-time info

93
149683 t/s$1.25 / $2.51M49.6May 2026
31

OpenAI · Default production tier

93
1489118 t/s$2 / $121M13.3Jul 2026
32

Anthropic · Balanced agents & coding value

93
147998 t/s$2 / $101M15.5Jun 2026
33

Z.ai (Zhipu AI) · Best open-weight price-to-performance

93
—117 t/s$0.15 / $0.51M286.2Aug 2026
34

Google · Fastest high-throughput generation

93
—345 t/s$0.75 / $3.751M41.3Sep 2026
35

Google · Multimodal + value

92
134587 t/s$1.25 / $101M16.4Mar 2025
36

Moonshot AI · Frontier quality at low cost

92
146648 t/s$0.73 / $3.49256K43.6Apr 2026
37

DeepSeek · Cheap open-weight agentic coding

92
—217 t/s$0.3 / $1.21M122.7Sep 2026
38

Anthropic · Complex analysis

91
136052 t/s$15 / $75200K2.0May 2025
39

DeepSeek · Cheap reasoning

91
135035 t/s$0.55 / $2.19128K66.4Jan 2025
40

Z.ai (Zhipu AI) · Fastest self-hostable frontier model

91
1483168 t/s$1.4 / $4.41M31.4Jun 2026
41

Anthropic · High-volume routing, extraction and subagent work

91
—255 t/s$0.1 / $0.51M303.3Oct 2026
42

Anthropic · Coding & balance

90
146773 t/s$3 / $151M10.0Feb 2026
43

Google · Fast multimodal at mid-tier price

90
1481148 t/s$0.75 / $3.751M40.0Jul 2026
44

OpenAI · Long context

89
1310120 t/s$2 / $81M17.8Apr 2025
45

DeepSeek · Cheap frontier-adjacent tokens (GA build)

89
146162 t/s$0.66 / $1.981M67.4Aug 2026
46

MiniMax · Open-weight agentic coding

89
145580 t/s$0.6 / $2.41M59.3Jun 2026
47

OpenAI · Cheapest current GPT-6 for high-volume work

89
——$0.1 / $0.51M296.7Sep 2026
48

OpenAI · Reasoning & math

88
1305155 t/s$1.1 / $4.4200K32.0Jan 2025
49

Anthropic · Coding & balance

88
132095 t/s$3 / $15200K9.8May 2025
50

Z.ai (Zhipu AI) · Open-weight agentic & tool use

88
146748 t/s$0.98 / $3.08200K43.3Apr 2026
51

xAI · Real-time info

87
133082 t/s$3 / $15131K9.7Feb 2025
52

Meta · Self-hosted multimodal agents (needs ~700GB GPU memory)

87
—172 t/sSelf-host1M, Oct 2026
53

Mistral AI · Open-weight coding agents on EU-hosted or self-run stacks

87
—120 t/s$0.7 / $2.11M62.1Oct 2026
54

DeepSeek · Best open-source value

86
131062 t/s$0.27 / $1.1128K125.5Mar 2025
55

Alibaba Cloud · Multilingual & APAC

86
1448124 t/s$1.4 / $5.6256K24.6Apr 2026
56

OpenAI · General purpose

85
1285109 t/s$2.5 / $10128K13.6May 2024
57

DeepSeek · Cheapest tokens, period

84
1441105 t/s$0.14 / $0.281M400.0Apr 2026
58
Beam NewOSS

Reflection AI · Apache-2.0 large open model for self-hosted coding agents

84
—287 t/sSelf-host1M, Oct 2026
59

OpenAI · High-volume cheap frontier-family

83
1436186 t/s$0.2 / $1.21M118.6Jul 2026
60

Meta · Open-source value

80
1260135 t/s$0.2 / $0.61M200.0Apr 2025
61

Alibaba Cloud · Open-source flagship

80
125585 t/s$0.3 / $0.9131K133.3Sep 2024
62

Alibaba Cloud · Locally-runnable open-weight (~17GB VRAM at 4-bit)

80
——Self-host262K, Aug 2026
63

Mistral AI · Multilingual

79
125078 t/s$2 / $6128K19.8Nov 2024
64

Google · High-volume cheap multimodal

79
1418192 t/s$0.3 / $2.51M56.4Jul 2026
65

NVIDIA · Efficient single-GPU deployment, 1M context

79
——Self-host1M, Aug 2026
66

xAI · Budget reasoning

78
1275165 t/s$0.3 / $0.5131K195.0Feb 2025
67

Perplexity · Search + citations

78
—65 t/s$3 / $15200K8.7Feb 2025
68

Mistral AI · Code generation

76
—195 t/s$0.3 / $0.9256K126.7Jan 2025
69

NVIDIA · Open multimodal

76
1361158 t/sSelf-host256K, Apr 2026
70

Tencent · Long-context retrieval & agentic search, no license restrictions

76
——Self-host256K, Jul 2026
71

Anthropic · Speed & cost

75
1230172 t/s$0.8 / $4200K31.3Oct 2024
72

Google · Self-hosted general purpose

75
1351142 t/sSelf-host128K, Apr 2026
73

Google · Fastest + cheapest

74
1240244 t/s$0.1 / $0.41M296.0Feb 2025
74

Alibaba Cloud · Open-source coding

74
—125 t/s$0.15 / $0.45131K246.7Nov 2024
75

Ant Group · Efficient MoE alternative to trillion-param flagships

74
——Self-host256K, Aug 2026
76

OpenAI · High throughput

72
1216183 t/s$0.15 / $0.6128K192.0Jul 2024
77

Meta · Longest context

71
1195198 t/s$0.15 / $0.410M258.2Apr 2025
78

Amazon · AWS ecosystem

70
—110 t/s$0.8 / $3.2300K35.0Dec 2024
79

Cohere · Enterprise RAG

68
117072 t/s$2.5 / $10128K10.9Aug 2024
Quality = composite benchmark (MMLU, HumanEval, MATH)Arena ELO = LMSYS Chatbot Arena ratingValue = quality per dollarPrice = input / output per 1M tokens

How We Rank AI Models

Our leaderboard uses a composite quality index that combines three key benchmarks: MMLU Pro (measuring knowledge and reasoning across 57 subjects), HumanEval (measuring code generation ability), and MATH (measuring mathematical problem-solving). Scores are normalized to a 0-100 scale and cross-referenced against LMSYS Chatbot Arena ELO ratings for real-world validation.

We track speed (tokens per second), time-to-first-token (TTFT), pricing, and context window size to give you a complete picture. The Value Score divides quality by cost, showing you which models deliver the most capability per dollar.

Key Trends in AI Model Performance

  • A new frontier #1: Anthropic's Claude Fable 5.2 (9 November 2026) tops the board on the Artificial Analysis Intelligence Index (65 against 60 for Gemini 4 Carbon at launch), and is priced at $8/$40, down from $10/$50 for Fable 5.1. The cost per task is the catch: roughly $27 per Index task at max effort, so Claude Opus 5.5 ($4/$20) remains Anthropic's own recommended default for most work
  • OpenAI split GPT-5.6 into three tiers: Sol ($5/$30), Terra ($2/$12), and Luna ($0.20/$1.20), GA 9 July. On 30 July OpenAI cut Terra 20% and Luna 80%, citing inference work that reduced serving cost 20% — the steepest cut of the year from a US lab, and a direct answer to the Chinese open-weight tier on price
  • Open weights have caught the frontier: Moonshot's Kimi K3 (2.8T MoE) ranked #3 overall on the Artificial Analysis Intelligence Index at its July 2026 launch — ahead of every proprietary model except Claude Fable 5 and GPT-5.6 Sol at the time — and took #1 on the Frontend Code Arena. It is not a lone outlier: DeepSeek V4 Pro posts 80.6% on SWE-bench Verified (matching Gemini 3.1 Pro) and MiniMax M3 80.5%, while GLM-5.2 scores 62.1% on SWE-bench Pro against GPT-5.5's 58.6% — an MIT-licensed model you can self-host outscoring a $5/$30 US flagship on agentic coding
  • "Open" now means datacentre-scale: the leading open models are trillion-parameter MoEs — GLM-5.2 needs ~1TB VRAM in BF16 (~8x H200 at FP8), Kimi K3 needs 64+ accelerators. The ecosystem has bifurcated into giant open MoEs and genuinely small models (Qwen 3.6 27B, Gemma 4, Nemotron 3 Nano Omni) that run on one GPU or a phone, with little in between
  • Meta left the open frontier: Llama 5 has not shipped and is now forecast for 2027. Meta pivoted to its first closed frontier model, Muse Spark (April 2026), with no weights and no architecture paper — leaving Chinese labs (DeepSeek, Moonshot, Z.ai, Alibaba, MiniMax) as the effective owners of the open-weight frontier
  • Reasoning is built in: Frontier models like GPT-5.6 Sol, Claude Opus 5, and Gemini 3.1 Pro ship extended thinking by default, while deep-research variants trade latency for accuracy on the hardest tasks
  • Million-token context is standard: 1M-token windows are now table stakes for flagship models across OpenAI, Anthropic, Google, and xAI
  • DeepSeek is the price floor: V4 Pro sits at $0.435/$0.87 per 1M tokens — launched at $1.74/$3.48, cut 75% as a promotion, then made permanent on 31 May 2026 — and V4 Flash at $0.14/$0.28 is the cheapest usable model anywhere, with cache hits at $0.0028. Measured cost-per-task is about $0.04 on V4 Pro against $0.94 on Kimi K3 and far more on the closed frontier. One dollar buys ~1.15M output tokens on V4 Pro, ~227K on GLM-5.2, ~67K on Kimi K3. Watch one thing: DeepSeek has announced 2x peak-hour pricing for two daily windows, not yet active as of 4 August

Choosing the Right Model

There is no single "best" model: it depends on your use case. For most applications, a model routing approach works best: route simple queries to fast, cheap models and complex queries to frontier models. This gives you the best of both worlds: low cost and high quality.