Every benchmark.
One leaderboard.
68 benchmarks from 19 independent publishers, reconciled into one set of standings every morning. BenchLeader runs no evaluations of its own: it lines up what Artificial Analysis, LMArena, Epoch AI, LiveBench, SWE-bench and the rest have already measured, so every model is judged on all of the evidence at once, at the reasoning effort it was actually run at.
- models ranked
- 357
- configurations ranked
- 610
- sources live
- 19/19
- GPT-6 Astra (max) input price moved from $10.00/M to $11.00/M (10% up).
- GPT-6 Astra (max) output price moved from $50.00/M to $55.00/M (10% up).
- GPT-6 Astra (max) moved from #2 to #1 in the BenchLeader Index.
- Claude Fable 5 moved from #6 to #2 in the BenchLeader Index.
- Read the daily digest
Standings9 Sept 2026
- 1
GPT-6 Astra maxOpenAI · $22.00/M70.6▼ 0.3 - 2
Claude Fable 5 Anthropic · $20.00/M70.4▲ 3.7 - 3Claude Fable 5.1 xhighAnthropic · $20.00/M70.3—
- 4Claude Opus 5 Anthropic · $10.00/M70.0—
- 5GPT-5.6 Sol highOpenAI · $8.00/M68.0▲ 6.9
- 6GPT-5.5 OpenAI · $11.25/M66.6▲ 5.1
- 7
Kimi K3 Moonshot AI · $6.00/M66.5▲ 5.0 - 8Claude Opus 4.7 Anthropic · $10.00/M65.5▲ 1.2
- 9
Muse Spark Meta65.4▲ 4.9 - 10Muse Spark 1.1 Meta · $2.00/M64.9▲ 4.7
Purple: leader. Arrows: change in index since the last time it moved. Full standings
The frontier board
The 10 models that matter this week, read three ways. Click a lane to re-order.
Model
- 1GPT-6 Astramax70.654$0.88
- 2Claude Fable 570.463$0.80
- 3Claude Opus 570.052$0.40
- 4GPT-5.6 Solhigh68.059$0.32
- 5Kimi K366.537$0.24
- 6Muse Spark 1.164.9194$0.09
- 7
GLM-5.3-Flash61.072$0.005 - 8
DeepSeek V4 Prohigh60.066$0.03 - 9Ling 3.0 Flash54.0290$0.006
- 10
Qwen3.5-4B47.127$0.002
Bars are relative to the best in the lane; purple is the lane leader.Cost is one agentic coding session, 60k tokens in and 4k out.Full quality boardFull speed boardFull cost board
Quality against…
Showing 14 models: the frontier no other model beats on both axes, the top eight by index, and the best model in each price band. Up and to the left is better: more quality for less money. Hover for details, click to open a model.
GPT-6 AstramaxClaude Fable 5Claude Opus 5GPT-5.6 SolhighGPT-5.5Kimi K3Claude Opus 4.7Muse Spark 1.1GLM-5.3-FlashDeepSeek V4 ProhighGPT-5.6 LunaxhighLing 3.0 FlashQwen3.5-4Bqwen3-4b-instruct-2507The BenchLeader Index
One row per model, showing its best-scoring reasoning effort; switch to all configurations to compare efforts. Click any column to sort. Methodology
357 of 357
| # | ||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 AstramaxOpenAI | 70.6±4 | $22.00 | 54 | 329 s | 1.1M | 2026-09-03 | 2026-09-08 |
| 2 | Claude Fable 5Anthropic | 70.4±3 | $20.00 | 63 | 92 s | 1M | 2026-06-09 | 2026-09-08 |
| 3 | Claude Fable 5.1xhighAnthropic | 70.3±7 | $20.00 | 58 | 132 s | 1M | 2026-09-01 | 2026-09-01 |
| 4 | Claude Opus 5Anthropic | 70.0±4 | $10.00 | 52 | 82 s | 1M | 2026-07-24 | 2026-09-05 |
| 5 | GPT-5.6 SolhighOpenAI | 68.0±6 | $8.00 | 59 | 31 s | 1M | 2026-07-09 | 2026-07-09 |
| 6 | GPT-5.5OpenAI | 66.6±4 | $11.25 | 88 | 62 s | 1.1M | 2026-04-23 | 2026-09-08 |
| 7 | Kimi K3Moonshot AIopen ↗ | 66.5±3 | $6.00 | 37 | 58 s | 1.0M | 2026-07-16 | 2026-09-05 |
| 8 | Claude Opus 4.7Anthropic | 65.5±3 | $10.00 | 46 | 21 s | 1M | 2026-04-16 | 2026-09-08 |
| 9 | Muse SparkMeta | 65.4±3 | – | – | – | 262k | 2026-04-08 | 2026-09-02 |
| 10 | Muse Spark 1.1Meta | 64.9±4 | $2.00 | 194 | 2.03 s | 1.0M | 2026-07-09 | 2026-09-08 |
| 11 | Claude Opus 4.8Anthropic | 64.7±3 | $10.00 | 58 | 29 s | 1M | 2026-05-28 | 2026-09-08 |
| 12 |  Gemini 3.7 FlashmediumGoogle | 64.4±4 | $1.50 | 282 | 5.27 s | 1M | 2026-08-13 | 2026-08-13 |
| 13 | GPT-5.6 TerraxhighOpenAI | 64.2±4 | $4.50 | 93 | 33 s | 1M | 2026-07-09 | 2026-09-08 |
| 14 | GPT-5.4 ProOpenAI | 64.2±4 | $67.50 | 1 | 6.70 s | 1.1M | 2026-03-05 | 2026-03-23 |
| 15 | Gemini 3.1 ProGoogle | 63.9±4 | $4.50 | 110 | 30 s | 1.0M | 2026-02-19 | 2026-09-08 |
| 16 | Qwen3 8maxAlibaba | 63.9±3 | $2.67 | 39 | 53 s | 1M | 2026-08-02 | 2026-09-08 |
| 17 | Gemini 3.5 FlashmediumGoogle | 63.9±3 | $3.38 | 209 | 17 s | 1M | 2026-05-19 | 2026-09-08 |
| 18 | Muse Spark 1.3xhighMeta | 63.9±5 | $2.00 | 193 | 40 s | 1M | 2026-09-02 | 2026-09-08 |
| 19 |  Grok 4.6mediumxAI | 63.8±7 | $3.00 | 55 | 37 s | 500k | 2026-08-12 | 2026-08-12 |
| 20 | Gemini 3.8 FlashmediumGoogle | 63.5±5 | $1.50 | 273 | 14 s | 1M | 2026-09-02 | 2026-09-02 |
| 21 | GPT-5.4OpenAI | 63.4±6 | $5.63 | 130 | 126 s | 1.1M | 2026-03-05 | 2026-09-08 |
| 22 | Grok 4.5xAI | 63.4±4 | $3.00 | 56 | 13 s | 500k | 2026-07-08 | 2026-09-08 |
| 23 | Claude Opus 4.6Anthropic | 63.1±4 | $10.00 | 38 | 2.05 s | 1M | 2026-02-05 | 2026-09-08 |
| 24 | Muse Spark 1.2Meta | 62.6±7 | $2.00 | 219 | 24 s | 1.0M | 2026-08-05 | 2026-08-09 |
| 25 | GPT-5.3 CodexOpenAI | 62.4±9 | $4.81 | 126 | 65 s | 400k | 2026-02-05 | 2026-09-08 |
| 26 | Qwen3 6maxAlibaba | 62.0±3 | $2.92 | 58 | 38 s | 256k | 2026-04-20 | – |
| 27 | Gemini 3 ProGoogle | 61.6±3 | $4.50 | – | – | 1.0M | 2025-11-18 | 2026-09-08 |
| 28 | GPT-5.2OpenAI | 61.5±4 | $4.81 | 72 | 138 s | 400k | 2025-12-11 | 2026-09-08 |
| 29 | Gemini 3 FlashthinkingGoogle | 61.5±3 | $1.13 | 200 | 6.12 s | 1M | 2025-12-17 | – |
| 30 | Claude Sonnet 5Anthropic | 61.2±3 | $4.00 | 79 | 193 s | 1M | 2026-06-30 | 2026-09-05 |
| 31 | GLM-5.3-FlashZhipu AIopen ↗ | 61.0±6 | $0.119 | 72 | 30 s | 1M | 2026-08-20 | 2026-09-08 |
| 32 | GPT-5.2 CodexOpenAI | 60.7±5 | $4.81 | 25 | 6.74 s | 400k | 2025-12-18 | 2026-09-08 |
| 33 | Gemini 3.6 FlashGoogle | 60.7±6 | $1.50 | 190 | 16 s | 1.0M | 2026-07-21 | 2026-07-21 |
| 34 | GPT-5 ProOpenAI | 60.7±6 | $41.25 | – | – | 400k | 2025-08-07 | 2025-11-13 · stale |
| 35 | Grok 4.3mediumxAI | 60.7±6 | $1.56 | 113 | 12 s | 1M | 2026-04-17 | – |
| 36 | Kimi K2.6Moonshot AIopen ↗ | 60.6±4 | $1.71 | 52 | 89 s | 262k | 2026-04-20 | 2026-09-08 |
| 37 | Claude Opus 4.5thinkingAnthropic | 60.6±3 | $10.00 | 46 | 18 s | 200k | 2025-11-24 | 2026-09-01 |
| 38 | Mimo v2 ProXiaomi | 60.5±4 | $3.00 | – | – | 1.0M | 2026-03-18 | 2026-09-08 |
| 39 | GPT-5.5 InstantOpenAI | 60.4±6 | $11.25 | 135 | 16 s | 400k | 2026-05-05 | 2026-09-02 |
| 40 | Claude Sonnet 4.6Anthropic | 60.3±3 | $6.00 | 43 | 2.12 s | 1M | 2026-02-17 | 2026-09-08 |
| 41 | o3OpenAI | 60.3±3 | $3.50 | 139 | 6.03 s | 200k | 2025-04-16 | 2026-09-02 |
| 42 | Deepseek v4 ProhighDeepSeek | 60.3±5 | – | – | – | – | 2026-04-24 | 2026-09-08 |
| 43 | GLM-5.1Zhipu AIopen ↗ | 60.1±3 | $2.15 | 60 | 66 s | 200k | 2026-04-07 | 2026-09-08 |
| 44 | DeepSeek V4 ProhighDeepSeekopen ↗ | 60.0±3 | $0.548 | 66 | 32 s | 1M | 2026-04-24 | 2026-09-08 |
| 45 | GLM-5.3maxZhipu AIopen ↗ | 59.9±3 | $2.15 | 60 | 36 s | 1M | 2026-08-14 | 2026-09-08 |
| 46 | Qwen3.7 PlusAlibaba | 59.7±7 | $1.13 | 56 | 38 s | 1M | 2026-06-02 | 2026-09-08 |
| 47 | GPT-5OpenAI | 59.5±4 | $3.44 | 76 | 74 s | 400k | 2025-08-07 | 2026-09-02 |
| 48 | GPT-5.1OpenAI | 59.4±5 | $3.44 | 105 | 54 s | 400k | 2025-11-12 | 2026-09-08 |
| 49 | GPT-5.6 LunaxhighOpenAI | 59.3±4 | $0.450 | 106 | 60 s | 1M | 2026-07-09 | 2026-09-08 |
| 50 | DeepSeek V4 FlashhighDeepSeekopen ↗ | 59.3±3 | $0.168 | 118 | 18 s | 1M | 2026-07-31 | 2026-09-08 |