One composite score across reasoning, coding, and agentic evals.
Median tokens per second across the APIs serving each model.
Dollars per million tokens on a 3:1 input-to-output blend.
Intelligence Index · best model per lab
Claude Fable 5 (with fallback)60
GPT-5.5 (xhigh)55
GLM-5.2 (max)51
Gemini 3.5 Flash50
Qwen3.7 Max46
MiniMax-M344
Muse Spark43
Nemotron 3 Ultra38
Grok 4.3 (high)38Median output tokens per second · best per lab
Qwen3.7 Max197GLM-5.2 (max)181Gemini 3.5 Flash174Nemotron 3 Ultra155
Mistral Medium 3.5130Grok 4.3 (high)126
Nova 2.0 Pro Preview (medium)125MiniMax-M391Claude Fable 5 (with fallback)73USD per 1M tokens, 3:1 blended · cheapest per lab
MiniMax-M3$0.53Nemotron 3 Ultra$1.18
MiMo-V2.5-Pro$1.35Grok 4.3 (high)$1.56
Kimi K2.6$1.71GLM-5.2 (max)$2.15Mistral Medium 3.5$3.00Gemini 3.5 Flash$3.38Nova 2.0 Pro Preview (medium)$3.44Every model we track, ranked by Intelligence Index.
| # | Model | Creator | Released | Context | Intelligence | Coding | Agentic | Speed (t/s) | Blended $/1M | Latency |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5 (with fallback) | Anthropic | Jun 2026 | 1M | 60 | 77 | 53 | 73 | $20.00 | 173s |
| 2 | Claude Opus 4.8 (max) | Anthropic | May 2026 | 1M | 56 | 74 | 47 | 65 | $10.00 | 18s |
| 3 | GPT-5.5 (xhigh) | OpenAI | Apr 2026 | 922k | 55 | 75 | 45 | 71 | $11.25 | 69s |
| 4 | Claude Opus 4.7 (max) | Anthropic | Apr 2026 | 1M | 54 | 74 | 44 | 48 | $10.00 | 18s |
| 5 | Claude Sonnet 5 (max) | Anthropic | Jun 2026 | 1M | 53 | 72 | 47 | 78 | $6.00 | 163s |
| 6 | GLM-5.2 (max)Open | Z AI | Jun 2026 | 1M | 51 | 69 | 43 | 181 | $2.15 | 13s |
| 7 | Gemini 3.5 Flash | May 2026 | 1M | 50 | 70 | 37 | 174 | $3.38 | 20s | |
| 8 | Claude Sonnet 4.6 (max) | Anthropic | Feb 2026 | 1M | 47 | 63 | 41 | 47 | $6.00 | 105s |
| 9 | Gemini 3.1 Pro Preview | Feb 2026 | 1M | 47 | 69 | 21 | 129 | $4.50 | 21s | |
| 10 | Qwen3.7 Max | Alibaba | May 2026 | 1M | 46 | 66 | 31 | 197 | $3.75 | 15s |
| 11 | MiniMax-M3Open | MiniMax | Jun 2026 | 1M | 44 | 59 | 35 | 91 | $0.53 | 24s |
| 12 | GPT-5.3 Codex (xhigh) | OpenAI | Feb 2026 | 400k | 44 | — | — | 84 | $4.81 | 112s |
| 13 | Muse Spark | Meta | Apr 2026 | 262k | 43 | 59 | 29 | — | — | — |
| 14 | GPT-5.4 mini (xhigh) | OpenAI | Mar 2026 | 400k | 40 | 56 | 30 | 170 | $1.69 | 6.9s |
| 15 | Qwen3.7 Plus | Alibaba | Jun 2026 | 1M | 39 | 56 | 21 | 50 | $0.59 | 43s |
| 16 | MiniMax-M2.7Open | MiniMax | Mar 2026 | 205k | 38 | 53 | 26 | 51 | $0.53 | 50s |
| 17 | Nemotron 3 UltraOpen | NVIDIA | Jun 2026 | 262k | 38 | 49 | 27 | 155 | $1.18 | 16s |
| 18 | Grok 4.3 (high) | xAI | Apr 2026 | 1M | 38 | 42 | 24 | 126 | $1.56 | 20s |
| 19 | GLM-5.1Open | Z AI | Apr 2026 | 200k | 35 | — | — | 51 | $2.15 | 1.8s |
| 20 | Kimi K2.6Open | Kimi | Apr 2026 | 256k | 35 | — | — | 46 | $1.71 | 2.8s |
| 21 | Qwen3.5 397B A17BOpen | Alibaba | Feb 2026 | 262k | 34 | 48 | 20 | 49 | $1.35 | 68s |
| 22 | Mistral Medium 3.5Open | Mistral | Apr 2026 | 256k | 30 | 47 | 19 | 130 | $3.00 | 17s |
| 23 | Claude 4.5 Haiku | Anthropic | Oct 2025 | 200k | 30 | 44 | 16 | 101 | $2.00 | 13s |
| 24 | Gemma 4 31BOpen | Apr 2026 | 256k | 29 | 43 | 14 | 35 | — | 51s | |
| 25 | MiMo-V2.5-ProOpen | Xiaomi | Apr 2026 | 1M | 28 | — | — | 54 | $1.35 | 2.3s |
| 26 | gpt-oss-120b (high)Open | OpenAI | Aug 2025 | 131k | 24 | 30 | 13 | 265 | $0.26 | 8.5s |
| 27 | Nova 2.0 Pro Preview (medium) | Amazon | Nov 2025 | 256k | 22 | 34 | 7 | 125 | $3.44 | 29s |
| 28 | MBZUAI Institute of Foundation Models | Dec 2025 | 262k | 17 | 21 | 2 | — | — | — | |
| 29 |  Solar Pro 3 | Upstage | Apr 2026 | 128k | 14 | 16 | 3 | — | — | — |
| — | GPT-5.5 Pro (xhigh) | OpenAI | Apr 2026 | 922k | — | — | — | — | — | — |
Composite of 10 evaluations spanning reasoning, knowledge, math, coding, and agentic tool use. Higher is better.
Incorporates GPQA Diamond, Humanity's Last Exam, AIME 2025, LiveCodeBench, SciCode, IFBench, Terminal-Bench Hard, τ²-Bench, and more.
Intelligence Index against blended USD per 1M tokens (3:1 input:output, log scale).
Up and to the left wins: more intelligence per dollar. Models without public API pricing are excluded.
Intelligence Index against median output tokens per second.
Up and to the right wins: smart and fast. Speed is the median across providers serving each model.
Intelligence Index by release date. The dashed line tracks the running frontier.
Claude Fable 5 set the current frontier on June 9, 2026 — 24 days before this snapshot.
Composite of coding evaluations (LiveCodeBench, SciCode, Terminal-Bench Hard). Higher is better.
Tool calling and long-horizon agent tasks (τ²-Bench, Terminal-Bench). Higher is better.
Individual evaluation scores (0–100) behind the Intelligence Index. Darker is better, normalized per column.
| Model | GPQA Diamond | Humanity's Last Exam | SciCode | IFBench | Terminal-Bench Hard | τ²-Bench Telecom | AA-LCR (Long Context) | CritPt | MMMU-Pro |
|---|---|---|---|---|---|---|---|---|---|
| Claude Fable 5 (with fallback) | 92.6 | 53.3 | 60.2 | 63.5 | 62.9 | 98.5 | 70.0 | 28.6 | — |
| Claude Opus 4.8 (max) | 92.0 | 45.7 | 53.5 | 62.2 | 58.3 | 94.4 | 67.7 | 20.9 | — |
| GPT-5.5 (xhigh) | 93.5 | 44.3 | 56.1 | 75.9 | 60.6 | 93.9 | 74.3 | 27.1 | 79.9 |
| Claude Opus 4.7 (max) | 91.4 | 39.6 | 54.5 | 58.6 | 51.5 | 88.6 | 70.3 | 12.0 | 78.8 |
| Claude Sonnet 5 (max) | 91.1 | 39.6 | 53.6 | — | — | — | 70.7 | 16.9 | 77.3 |
| GLM-5.2 (max) | 89.5 | 40.1 | 50.5 | 73.3 | 50.8 | 99.1 | 71.3 | 20.9 | — |
| Gemini 3.5 Flash | 92.2 | 41.0 | 53.1 | 76.3 | 40.9 | 95.3 | 69.3 | 13.1 | 84.3 |
| Claude Sonnet 4.6 (max) | 87.5 | 30.0 | 46.8 | 56.6 | 53.0 | 75.7 | 70.7 | 3.1 | 73.3 |
| Gemini 3.1 Pro Preview | 94.1 | 44.7 | 58.9 | 77.1 | 53.8 | 95.6 | 72.7 | 17.7 | 82.4 |
| Qwen3.7 Max | 92.3 | 38.1 | 48.8 | 80.5 | 50.8 | 94.7 | 69.0 | 13.4 | — |
| MiniMax-M3 | 92.9 | 37.1 | 45.4 | 82.9 | 42.4 | 88.9 | 74.0 | 3.7 | 78.6 |
| GPT-5.3 Codex (xhigh) | 91.5 | 39.9 | 53.2 | 75.4 | 53.0 | 86.0 | 74.0 | 16.9 | 78.5 |
| Muse Spark | 88.4 | 39.9 | 51.5 | 75.9 | 45.5 | 91.5 | 69.7 | 11.3 | 80.5 |
| GPT-5.4 mini (xhigh) | 87.5 | 26.6 | 49.9 | 73.3 | 52.3 | 83.3 | 69.3 | 10.0 | 73.3 |
| Qwen3.7 Plus | 90.0 | 33.4 | 45.5 | 78.0 | 47.0 | 93.0 | 65.0 | 9.1 | 80.5 |
| MiniMax-M2.7 | 87.4 | 28.1 | 47.0 | 75.7 | 39.4 | 84.8 | 68.7 | 0.6 | — |
| Nemotron 3 Ultra | 86.7 | 26.6 | 39.9 | 81.4 | 36.4 | 83.3 | 67.0 | 3.1 | — |
| Grok 4.3 (high) | 90.1 | 35.0 | 47.3 | 81.3 | 37.9 | 97.7 | 64.3 | 8.0 | 78.1 |
| GLM-5.1 | 83.9 | 25.6 | 36.1 | 52.0 | 35.6 | 97.1 | 44.3 | 0.0 | — |
| Kimi K2.6 | 78.8 | 18.2 | 39.5 | 44.3 | 37.9 | 93.9 | 57.7 | 1.4 | — |
| Qwen3.5 397B A17B | 89.3 | 27.3 | 42.0 | 78.8 | 40.9 | 95.6 | 65.7 | 1.7 | 77.3 |
| Mistral Medium 3.5 | 74.8 | 12.8 | 39.6 | 68.8 | 33.3 | 94.2 | 61.0 | 0.0 | 64.9 |
| Claude 4.5 Haiku | 67.2 | 9.7 | 43.3 | 54.3 | 27.3 | 54.7 | 70.3 | 0.0 | 58.6 |
| Gemma 4 31B | 85.7 | 22.7 | 43.4 | 75.6 | 36.4 | 59.9 | 62.0 | 1.4 | 73.4 |
| MiMo-V2.5-Pro | 76.2 | 13.3 | 39.1 | 42.7 | 35.6 | 72.5 | 35.0 | 1.1 | — |
| gpt-oss-120b (high) | 78.2 | 18.5 | 38.9 | 69.0 | 23.5 | 65.8 | 50.7 | 1.1 | — |
| Nova 2.0 Pro Preview (medium) | 78.5 | 8.9 | 42.7 | 79.0 | 24.2 | 92.7 | 54.3 | 0.0 | 64.5 |
| 71.3 | 9.5 | 33.0 | 62.8 | 6.8 | 25.4 | 52.7 | 0.0 | — | |
| Solar Pro 3 | 72.4 | 10.1 | 24.7 | 71.2 | 7.6 | 86.3 | 27.0 | 0.0 | — |
| GPT-5.5 Pro (xhigh) | — | — | — | — | — | — | — | 30.6 | — |
AIME 2025 and LiveCodeBench are retired for newer models and excluded here; MMMU-Pro applies to multimodal-evaluated models only.
Knowledge reliability from -100 to 100: correct answers score positive, hallucinated ones negative.
A negative score means the model hallucinates more than it knows. Declining to answer scores zero — most models would rather guess.
Elo from blind pairwise comparisons on real economically valuable work tasks, with web and shell access.
Higher is better. Judged across occupations from software engineering to financial analysis.
Average precision at full recall diagnosing live Kubernetes incidents. Higher is better.
Models investigate real cluster telemetry to find root causes. Even the frontier tops out below 0.5 — ops work is far from solved.
USD to complete every evaluation in the Intelligence Index, including reasoning tokens. Lower is better.
The spread is real: the same suite costs $19 on gpt-oss-20B and over $4,600 on the priciest frontier models.
Median output tokens per second across providers serving each model. Higher is better.
Seconds from request to first answer token, including reasoning time. Lower is better.
Max-effort reasoning modes pay for their scores in wait time: the smartest configurations routinely think for one to two minutes.
USD per 1M tokens by direction. Lower is better.
Output tokens typically cost 2–4× input. Reasoning tokens bill as output, so thinking models multiply effective price.
Maximum input tokens per request.
Weights availability plus transparency of methodology and training data, 0–100.
Only models with published openness scores shown. K2 Think V2 and Nemotron 3 Ultra lead; most frontier labs publish nothing.
Reporting from the eval desk
By DCD · 2 hrs ago
By PlasticsToday · 2 hrs ago
By FinanzNachrichten.de · 2 hrs ago
By Legit.ng · 2 hrs ago
Methodology: indices are composites of public evaluations run independently with standardized prompts; speed and latency are medians measured across API providers over the trailing 72 hours. Benchmark data: public model-evaluation snapshot, July 3, 2026. Prices are list API prices and change frequently. Company logos identify the respective model creators.