Model selection · Independent benchmarks
Compare the models. Then test your work.
Compare measured intelligence, coding and agentic capability alongside API token costs. These general benchmarks help build a shortlist; they do not measure CRE underwriting accuracy.
Artificial Analysis via OpenRouter
Capability and cost, side by side.
Source: Artificial Analysis (artificialanalysis.ai) via OpenRouter (openrouter.ai/rankings).
Snapshot as of . The source snapshot timestamp does not identify when every model was evaluated.
Licensed under CC BY 4.0. View the license. Artificial Analysis source.
Showing 100 of 100 models in this snapshot. Intelligence scores are a separate index, not a percentage.
| Model / configuration | Intelligence index | Input / 1M | Output / 1M | Estimated token cost |
|---|---|---|---|---|
| Claude Opus 5.5 (Max, Default Fallback)anthropic/claude-opus-5.5-20260921 | 57.6 | $4.40 | $22.00 | $9.90 |
| Claude Sonnet 5.5 (Max, Default Fallback)anthropic/claude-sonnet-5.5-20260928 | 56.0 | $2.00 | $10.00 | $4.50 |
| Claude Fable 5.1 (Max, Default Fallback)anthropic/claude-fable-5.1-20260831 | 53.4 | $5.00 | $25.00 | $11.25 |
| Qwen3.8 Maxqwen/qwen3.8-max-20260803 | 53.4 | Not reported | Not reported | Not reported |
| GPT-6 Astra (Max)openai/gpt-6-astra-20260903 | 52.7 | $60.00 | $300.00 | $135.00 |
| Gemini 4 Argon (High)google/gemini-4-argon-20261013 | 52.6 | $2.20 | $11.00 | $4.95 |
| GPT-6.1 Sol (Max)openai/gpt-6.1-sol-20260929 | 51.8 | $12.00 | $60.00 | $27.00 |
| Claude Opus 5 (Max)anthropic/claude-opus-5-20260723 | 50.8 | $5.50 | $27.50 | $12.38 |
| Claude Fable 5 (Max, Opus 4.8 Fallback)anthropic/claude-5-fable-20260609 | 49.6 | $10.00 | $50.00 | $22.50 |
| Muse Spark 1.3 (Max)meta/muse-spark-1.3-20260902 | 48.1 | $1.25 | $4.25 | $2.31 |
| GPT-6 Sol (Max)openai/gpt-6-sol-20260922 | 47.6 | $2.00 | $10.00 | $4.50 |
| GPT-5.6 Sol (Max)openai/gpt-5.6-sol-20260709 | 47.0 | $8.00 | $40.00 | $18.00 |
| Grok 4.7 (Xhigh)x-ai/grok-4.7-20260916 | 46.4 | $2.20 | $6.60 | $3.85 |
| MiMo-V2.6-Proxiaomi/mimo-v2.6-pro-20260921 | 46.3 | $0.44 | $0.87 | $0.65 |
| Qwen3.8 Max (0902)qwen/qwen3.8-max-20260902 | 45.4 | $2.00 | $6.00 | $3.50 |
| GLM-5.3 (Max)z-ai/glm-5.3-20260816 | 44.8 | $0.08 | $4.38 | $1.18 |
| Grok 4.6 (High)x-ai/grok-4.6-20260810 | 44.3 | $2.20 | $6.60 | $3.85 |
| Step 5 Previewstepfun/step-5-preview-20261008 | 43.7 | $1.00 | $2.70 | $1.68 |
| Kimi K3 (Max)moonshotai/kimi-k3-20260715 | 43.6 | $0.24 | $11.94 | $3.22 |
| Claude Haiku 5.5 (Max)anthropic/claude-haiku-5.5-20261007 | 43.4 | $0.11 | $0.55 | $0.25 |
| MiniMax-M3.1-Flashminimax/minimax-m3.1-flash-20261019 | 42.7 | $0.00 | $0.00 | $0.00 |
| GPT-5.6 Terra (Max)openai/gpt-5.6-terra-20260709 | 42.1 | $4.00 | $24.00 | $10.00 |
| Claude Opus 4.8 (Max)anthropic/claude-4.8-opus-20260528 | 41.8 | $10.00 | $50.00 | $22.50 |
| GLM-5.3-Flashz-ai/glm-5.3-flash-20260826 | 41.8 | $0.12 | $0.39 | $0.22 |
| Ling 3.1 Flashinclusionai/ling-3.1-flash-20261002 | 41.1 | $0.08 | $0.22 | $0.13 |
| Gemini 3.8 Flash (High)google/gemini-3.8-flash-20260902 | 40.9 | $0.83 | $4.13 | $1.86 |
| Claude Opus 4.7 (Max)anthropic/claude-4.7-opus-20260416 | 40.7 | $5.00 | $25.00 | $11.25 |
| Qwen3.8 2.4T A95Bqwen/qwen3.8-2.4t-a95b-20260812 | 39.9 | $2.00 | $6.00 | $3.50 |
| Gemini 3.7 Flash (Medium)google/gemini-3.7-flash-20260813 | 39.6 | $0.83 | $4.13 | $1.86 |
| Muse Spark 1.2 (Xhigh)meta/muse-spark-1.2-20260805 | 39.6 | $1.25 | $4.25 | $2.31 |
| DeepSeek V4.1 Flash (Max)deepseek/deepseek-v4.1-flash-20261017 | 39.5 | $0.08 | $0.50 | $0.21 |
| GPT-5.4 (Xhigh)openai/gpt-5.4-20260305 | 39.0 | $5.00 | $30.00 | $12.50 |
| Grok 4.5 (High)x-ai/grok-4.5-20260708 | 38.8 | $4.00 | $12.00 | $7.00 |
| Mistral Large 4 Previewmistralai/mistral-large-4-0-20261006 | 38.4 | $1.50 | $4.60 | $2.65 |
| GPT-5.5 (Xhigh)openai/gpt-5.5-20260423 | 38.4 | $2.50 | $15.00 | $6.25 |
| Claude Sonnet 5 (Max)anthropic/claude-sonnet-5-20260630 | 38.2 | $2.20 | $11.00 | $4.95 |
| GPT-6 Luna (Max)openai/gpt-6-luna-20260922 | 38.1 | $0.10 | $0.50 | $0.23 |
| MiMo-V2.6-Flashxiaomi/mimo-v2.6-flash-20260921 | 37.9 | $0.13 | $0.28 | $0.20 |
| GPT-5.6 Luna (Max)openai/gpt-5.6-luna-20260709 | 37.3 | $0.40 | $2.40 | $1.00 |
| DeepSeek V4 Pro 0813 (Max)deepseek/deepseek-v4-pro-20260813 | 36.0 | $0.88 | $3.80 | $1.83 |
| DeepSeek V4 Flash Vision (Max)deepseek/deepseek-v4-flash-vision-exp-20260821 | 34.8 | $0.44 | $1.32 | $0.77 |
| DeepSeek V4 Flash 0731 (Max)deepseek/deepseek-v4-flash-20260731 | 34.3 | $0.19 | $0.56 | $0.33 |
| Gemini 3.6 Flash (High)google/gemini-3.6-flash-20260721 | 34.0 | $1.35 | $6.75 | $3.04 |
| Muse Spark 1.1 (Xhigh)meta/muse-spark-1.1-20260709 | 33.7 | $1.25 | $4.25 | $2.31 |
| Qwen3.8 27B (Xhigh)qwen/qwen3.8-27b-20260814 | 33.7 | $0.04 | $2.30 | $0.62 |
| GLM-5.2 (Max)z-ai/glm-5.2-20260616 | 33.7 | $0.45 | $1.44 | $0.81 |
| Gemini 3.5 Flash (Medium)google/gemini-3.5-flash-20260519 | 33.6 | $2.70 | $16.20 | $6.75 |
| GPT-5.3 Codex (Xhigh)openai/gpt-5.3-codex-20260224 | 32.5 | $3.50 | $28.00 | $10.50 |
| Claude Opus 4.6 (Max)anthropic/claude-4.6-opus-20260205 | 31.9 | $5.00 | $25.00 | $11.25 |
| DeepSeek V4 Pro 0424 (Max)deepseek/deepseek-v4-pro-20260423 | 30.4 | $1.56 | $3.12 | $2.34 |
| GPT-5.2 (Xhigh)openai/gpt-5.2-20251211 | 30.4 | $3.50 | $28.00 | $10.50 |
| Claude Sonnet 4.6 (Max)anthropic/claude-4.6-sonnet-20260217 | 30.1 | $3.00 | $15.00 | $6.75 |
| Gemini 3.1 Pro Previewgoogle/gemini-3.1-pro-preview-20260219 | 29.7 | $3.60 | $21.60 | $9.00 |
| Qwen3.7 Maxqwen/qwen3.7-max-20260520 | 29.5 | $2.00 | $6.00 | $3.50 |
| MiniMax-M3minimax/minimax-m3-20260531 | 29.2 | $0.22 | $0.88 | $0.44 |
| Claude Opus 4.5 (Reasoning)anthropic/claude-4.5-opus-20251124 | 29.1 | $5.00 | $25.00 | $11.25 |
| MiMo-V2-Proxiaomi/mimo-v2-pro-20260318 | 28.6 | Not reported | Not reported | Not reported |
| GPT-5.2 Codex (Xhigh)openai/gpt-5.2-codex-20260114 | 28.5 | $1.75 | $14.00 | $5.25 |
| Qwen3.6 Max Previewqwen/qwen3.6-max-preview-20260420 | 28.4 | Not reported | Not reported | Not reported |
| Nex-N2-Pro (Based on Qwen3.5-397B-A17B)nex-agi/nex-n2-pro | 28.2 | Not reported | Not reported | Not reported |
| Solar Pro 4upstage/solar-pro4-20260810 | 28.2 | $0.15 | $0.60 | $0.30 |
| GLM-5 (Reasoning)z-ai/glm-5-20260211 | 27.9 | $1.00 | $3.20 | $1.80 |
| Grok Build 0.1 0616x-ai/grok-build-0.1-20260520 | 27.2 | $2.00 | $4.00 | $3.00 |
| Kimi K2.6 (Reasoning)moonshotai/kimi-k2.6-20260420 | 27.0 | $0.86 | $3.60 | $1.76 |
| Qwen3.6 Plusqwen/qwen3.6-plus-04-02 | 27.0 | $0.33 | $1.95 | $0.81 |
| GLM-5-Turboz-ai/glm-5-turbo-20260315 | 26.6 | $1.20 | $4.00 | $2.20 |
| Nemotron 3.5 Super VLnvidia/nemotron-3.5-super-vl-20261015 | 26.5 | $0.25 | $1.25 | $0.56 |
| Gemini 3 Flash Preview (Reasoning)google/gemini-3-flash-preview-20251217 | 26.3 | $0.90 | $5.40 | $2.25 |
| GLM-5.1 (Reasoning)z-ai/glm-5.1-20260406 | 26.1 | $1.33 | $4.18 | $2.38 |
| MiMo-V2.5-Pro (Reasoning)xiaomi/mimo-v2.5-pro-20260422 | 26.0 | $0.30 | $0.61 | $0.46 |
| Kimi K2.7 Codemoonshotai/kimi-k2.7-code-20260612 | 25.8 | $0.95 | $4.00 | $1.95 |
| Inkling Smallthinkingmachines/inkling-small-20260730 | 25.7 | $0.00 | $0.00 | $0.00 |
| Grok 4.20 0309 v2 (Reasoning)x-ai/grok-4.20-20260309 | 25.7 | $2.50 | $5.00 | $3.75 |
| Hy3tencent/hy3-preview-20260421 | 25.3 | $0.18 | $0.60 | $0.33 |
| Qwen3.7 Plusqwen/qwen3.7-plus-20260602 | 25.2 | $0.23 | $0.92 | $0.46 |
| MiMo-V2-Omni-0327xiaomi/mimo-v2-omni-20260318 | 25.1 | Not reported | Not reported | Not reported |
| Inkling (Xhigh)thinkingmachines/inkling-20260715 | 25.0 | $0.00 | $0.00 | $0.00 |
| GPT-5 Codex (High)openai/gpt-5-codex | 24.9 | Not reported | Not reported | Not reported |
| Grok 4.3 (High)x-ai/grok-4.3-20260430 | 24.9 | $1.00 | $2.00 | $1.50 |
| DeepSeek V4.1 Flash (Non-reasoning)deepseek/deepseek-v4.1-flash-20260910 | 24.7 | $0.30 | $1.20 | $0.60 |
| GPT-5.1 (High)openai/gpt-5.1-20251113 | 24.7 | $2.50 | $20.00 | $7.50 |
| Ling-3.0-flash-VLinclusionai/ling-3.0-flash-vl-20260910 | 24.6 | $0.06 | $0.18 | $0.11 |
| DeepSeek V4 Flash 0420 (High)deepseek/deepseek-v4-flash-20260423 | 24.4 | $0.15 | $0.30 | $0.23 |
| GPT-5.4 mini (Xhigh)openai/gpt-5.4-mini-20260317 | 24.1 | $1.50 | $9.00 | $3.75 |
| Solar Mini 4upstage/solar-mini4-20260922 | 24.1 | $0.05 | $0.20 | $0.10 |
| GPT-5.1 Codex (High)openai/gpt-5.1-codex-20251113 | 23.7 | $1.25 | $10.00 | $3.75 |
| Kimi K2.5 (Reasoning)moonshotai/kimi-k2.5-0127 | 23.5 | $0.60 | $3.00 | $1.35 |
| GLM 5V Turbo (Reasoning)z-ai/glm-5v-turbo-20260401 | 23.5 | $1.20 | $4.00 | $2.20 |
| GPT-5 (High)openai/gpt-5-2025-08-07 | 23.0 | $0.63 | $5.00 | $1.88 |
| Nemotron 3 Ultra 550B A55B (Reasoning)nvidia/nemotron-3-ultra-550b-a55b-20260604 | 22.9 | $0.40 | $1.76 | $0.84 |
| Qwen3.5 27B (Reasoning)qwen/qwen3.5-27b-20260224 | 22.9 | $0.25 | $2.00 | $0.75 |
| Claude 4.1 Opus (Reasoning)anthropic/claude-4.1-opus-20250805 | 22.8 | $7.50 | $37.50 | $16.88 |
| MiniMax-M2.5minimax/minimax-m2.5-20260211 | 22.8 | $0.30 | $1.20 | $0.60 |
| MiniMax-M2.7minimax/minimax-m2.7-20260318 | 22.8 | $0.30 | $1.20 | $0.60 |
| Ling-3.0-flash-Fininclusionai/ling-3.0-flash-fin-20260827 | 22.6 | Not reported | Not reported | Not reported |
| Grok 4x-ai/grok-4-07-09 | 22.5 | Not reported | Not reported | Not reported |
| MiMo-V2-Flash (Feb 2026)xiaomi/mimo-v2-flash-20251210 | 22.4 | Not reported | Not reported | Not reported |
| Gemini 3.5 Flash-Litegoogle/gemini-3.5-flash-lite-20260721 | 22.2 | $0.54 | $4.50 | $1.67 |
| GLM-4.7 (Reasoning)z-ai/glm-4.7-20251222 | 22.2 | $0.32 | $1.40 | $0.67 |
| Kimi K2 Thinkingmoonshotai/kimi-k2-thinking-20251106 | 22.0 | $0.60 | $2.50 | $1.23 |
Missing scores and prices stay “Not reported.” The feed returns up to 100 models. Results reflect the source's model and configuration labels; coverage and evaluation settings can differ.
Use a benchmark to choose what to test.
Keep each index separate
Each view uses one Artificial Analysis index. Intelligence, coding and agentic scores measure different evaluations. They are not percentages and are not combined with popularity, Arena ratings or individual test accuracies.
Compare the same workload
Use the same input and output quantities for each model. The estimate uses the published prompt and completion prices. Actual costs depend on the tokens consumed and your provider's rates.
Validate on your own CRE files
Run a shortlist on the same rent roll, T-12 and lease examples. Compare extraction accuracy, arithmetic, cited evidence and reviewer time against known answers before choosing a model for live work.