Top pick
#2 on the Index of the 250 models that fit: the one above it costs at least 4× as much.
- Index
- 68.9±1.9
- A month at your volume
- $106
- Per 1M at your mix
- $2.94
- Output speed
- 64 tok/s
- First answer
- 73 s
- Context
- 1.0M
Say how much quality, speed and cost matter, what the model has to be good at, and what you cannot do without. The answer updates as you go: a top pick, the runners-up, why each is where it is, and what a cheaper or faster choice would cost you in quality. Every score comes from independent benchmarks; nothing a model's maker published about itself is used.
331 ranked models · list prices and speeds · data as of 16 Nov 2026
Quality first, cost second, speed a tiebreak; about 36M tokens a month.
250 of 331 ranked models fit. Ranked on the Index (57%), speed (14%) and cost (29%) at 36M tokens a month.
Top pick
#2 on the Index of the 250 models that fit: the one above it costs at least 4× as much.
#23 on the Index, 6.7 points behind the top pick, at half the top pick's price, 3× the output speed.
#24 on the Index, 6.8 points behind the top pick, 2.6× cheaper than the top pick, 1.5× the output speed.
#1 on the Index, 3.8 points ahead of the top pick, 1.7× the output speed, 4× the price of the top pick.
#27 on the Index, 7.3 points behind the top pick, at half the top pick's price, 2.5× the output speed.
| # | Model | Index | A month | Speed | First answer | Context |
|---|---|---|---|---|---|---|
| 1 | Gemini 4 Carbonhigh | 68.9 | $106 | 64 | 73 s | 1.0M |
| 2 | Muse Spark 1.1 | 62.2 | $55 | 193 | 1.8 s | 1.0M |
| 3 | Gemini 3.7 Flashhigh | 62.1 | $40 | 95 | 2.2 s | 1.0M |
| 4 | Claude Fable 5.2max | 72.7 | $421 | 109 | 4 min | 1M |
| 5 | Muse Spark 1.3thin | 61.6 | $55 | 163 | 2.6 s | 1.0M |
| 6 | Gemini 4 Hydrogenhigh | 63.4 | $79 | 170 | 16 s | 1.0M |
| 7 | Gemini 3.6 Flashhigh | 61.0 | $40 | 177 | 11 s | 1.0M |
| 8 | DeepSeek V4 Flashhigh | 57.2 | $7.0 | 82 | 0.7 s | 1.0M |
| 9 | Muse Spark 1.2xhigh | 61.5 | $55 | 208 | 11 s | 1.0M |
| 10 | Qwen3.6 Maxmax | 62.1 | $77 | – | – | 262k |
| 11 | Gemini 3 Flashthinkingthin | 58.9 | $30 | 196 | 4.7 s | 1.0M |
| 12 | Gemini 3.5 Flashhigh | 62.5 | $89 | 176 | 9.9 s | 1.0M |
| 13 | GPT-6.1 Terramax | 64.1 | $53 | 72 | 3 min | 1.1M |
| 14 | Gemma 4 26B A4B | 56.1 | $7.1 | 76 | 0.4 s | 262k |
| 15 | GPT-5 nanothin | 56.2 | $3.6 | 106 | 2.9 s | 400k |
| 16 | Claude Opus 5.5max | 68.4 | $212 | 116 | 6 min | 1M |
| 17 | Claude Haiku 5.5xhigh | 56.7 | $5.3 | 231 | 50 s | 1M |
| 18 | Gemini 4 Argonhigh | 65.3 | $106 | 66 | 46 s | 1.0M |
| 19 | GPT-6.1 Solmax | 66.6 | $106 | 50 | 3 min | 1.1M |
| 20 | GPT-5.6 Lunaxhigh | 58.2 | $12 | 110 | 28 s | 1.1M |
| 21 | Grok 4.20 Multi-Agentthin | 58.5 | $45 | 349 | 12 s | 2M |
| 22 | Gemini 3.8 Flashmedium | 60.2 | $40 | 116 | 15 s | 1.0M |
| 23 | DeepSeek V4.1 Flash 1017max | 57.7 | $14 | 174 | 40 s | 1.0M |
| 24 | Beam 1.0 501B A23B | 54.2 | $13 | 193 | 0.3 s | 1.0M |
| 25 | Step 3.5 Flash | 56.3 | $4.2 | 133 | 25 s | 256k |
| 26 | Claude Sonnet 5.5max | 64.2 | $106 | 152 | 5 min | 1M |
| 27 | Qwen3.5 Plusthinking | 57.4 | $26 | 80 | 0.8 s | 1M |
| 28 | Grok 4.1 Fastthinking | 55.3 | $7.9 | – | – | 128k |
| 29 | Inkling 2 Small | 57.9 | $13 | 100 | 49 s | 1.0M |
| 30 | GPT-6 Solmax | 62.6 | $106 | 127 | 37 s | 1.1M |
| 31 | Muse Wonderthin | 64.4 | $212 | 70 | 3.3 s | 1.0M |
| 32 | GPT-5.6 Solmax | 65.3 | $213 | 100 | 34 s | 1.1M |
| 33 | MiMo-V2.5 | 56.3 | $3.8 | 28 | 2.5 s | 1M |
| 34 | GLM 5.3 Flash | 57.8 | $6.6 | 54 | 87 s | 1M |
| 35 | GLM-5thinking | 60.2 | $43 | 69 | 29 s | 205k |
| 36 | Muse Embermax | 52.1 | $5.3 | 437 | 5.0 s | 1.0M |
| 37 | o3 | 60.4 | $97 | 131 | 5.2 s | 200k |
| 38 | GPT-5 mini | 56.6 | $18 | 70 | 3.5 s | 400k |
| 39 | MiMo-V2-Flashthinking | 53.9 | $4.8 | – | – | 256k |
| 40 | Nemotron 3 Ultra 550B A55B | 55.4 | $32 | 106 | 0.3 s | 1M |
Requirements come first and are absolute: a model over your budget, slower to its first answer than you allow, or without a result on image input when you need one, is out. What is left is ranked on three things. Quality is the mean of the capabilities you picked (category scores of the BenchLeader Index, where 50 is the average evaluated model and 15 points is one standard deviation). Speed is output tokens per second and the wait for the first answer. Cost is what a month at your volume comes to on list prices, input, cached input and output each at its own rate.
Each is measured in standard deviations across the models that fit, with speed and cost on a log scale, and the sliders weigh them: each step up doubles the weight. A bill under about $20 a month counts as cheap whatever the model, so pennies never outvote quality. Each model is represented by its best reasoning-effort setting for your weights, and a model needs most of the weighting measured, and a price when cost counts, to be ranked at all.
The same weighting, fixed and printed, is behind the best model by use case pages. To filter without weighting, use find a model; to price one model across plans and usage patterns, the cost calculator.