AI model comparison — text, image and video
The leading AI models from the US, Europe and China — text, image and video generation — compared by quality (market benchmarks), cost and open-source status, always separating remote (API) and local (on your hardware) models.
Data as of 2026-09-04 · automated research (Artificial Analysis, LMArena, official pricing) — verify before deciding.
📊 How quality is measured — three indices
We show quality three complementary ways. Here is how each index is built before you read the charts:
① Zendoric Quality (0-100) = equal-thirds average of SWE-bench-Pro (33% · real software development, checked against the maker) + LMArena (33% · human preference, normalised Elo) + Terminal-Bench (33% · agentic terminal capability). If a model lacks one of the three, its weight is shared among those present (at least two required).
② AA Index (Artificial Analysis Intelligence Index, 0-100) = a broader composite index (reasoning, science, code, maths). It gives a second reading: depending on how you measure, the maker ranking changes.
③ Cybersecurity (0-100) = capability on expert cyber tasks (hard «unguided pass@1» protocol: vuln-research and realistic exploitation). We use a non-saturated metric (the top is around 71, not 100, leaving headroom), not the Cybench «pass@k» the frontier already saturates. Sources: UK AISI, NIST-CAISI, CVE-Bench. We frame it as capability and risk, not an offensive ranking; where there is no direct eval it is estimated «est.».
📈 Zendoric Quality over time (frontier makers)
Quality index (0-100) of the top makers (their best model), last 24 months. Dashed line = quality estimated from the AA Index (labs without SWE-bench-Pro). Updated daily.
📈 AA Index over time (frontier makers)
AA Index (Artificial Analysis Intelligence Index, 0-100) of the top makers (their best model), last 24 months. It is a broader composite index (reasoning, science, code, maths) than ours. The historical series is reconstructed by anchoring each maker's trajectory to its current AA. Updated daily.
🛡️ Cybersecurity over time (frontier makers)
Cybersecurity index (0-100) of each maker's best model, last 24 months. Metric: EXPERT cyber tasks under a hard «unguided pass@1» protocol (no hints, one attempt; vuln-research and realistic exploitation). We pick it because it is NOT saturated — the top is around 71, not 100, so it discriminates and shows headroom (we drop Cybench «pass@k», where the frontier already scores ~100%). Sources: UK AISI (GPT-5.5 71.4% vs Anthropic preview 68.6%), NIST-CAISI, CVE-Bench. High confidence only for OpenAI/Anthropic (measured by AISI); the rest imputed by proximity → the whole series is marked «est.». We frame it as capability and RISK to govern, not an offensive ranking. Updated daily.
💰 Zendoric Quality vs cost
Flagship models of the top makers by quality (a maker may have several, e.g. Anthropic: Opus 5, Opus 4.8 and Fable 5). HIGHER = more quality; LEFT = cheaper (log axis). Hollow dot = quality estimated (AA Index). Colour by maker.
💰 AA Index vs cost
Same format as the quality/cost chart, but the vertical axis is the AA Index. HIGHER = more capability; LEFT = cheaper (log axis). Hollow dot = estimated AA (Terminal-Bench/SWE-Pro). Colour by maker.
🏁 Efficient frontier — quality vs cost
Every cloud model with data. Y = AA Index; X = output cost, log and REVERSED: further RIGHT is cheaper — the top-right corner is ideal. The green line joins the efficient frontier: models with no alternative that is both better and cheaper; picking off the line is only justified by non-price factors (ecosystem, context, open source). Hollow dot = estimated AA. Hover a dot for details.
🏆 Zendoric Quality (SW dev + arena + agentic)
| Model | Zendoric Quality | SWE-bench-Pro | DeepSWE | LMArena | Terminal-Bench | LiveCodeBench | GPQA | ARC-AGI-2 |
|---|---|---|---|---|---|---|---|---|
| 🇨🇳 Kimi K3Moonshot AI · China | 91.4 | — | 68.5 | 1489 | 88.3 | 68.0 | 93.5 | — |
| 🆕 🇺🇸 Claude Fable 5.1Anthropic · USA | 90.6 | 81.2 | — | 1504 | — | — | — | — |
| 🇺🇸 Claude Fable 5Anthropic · USA | 90.1 | 80.3 | 69.7 | 1515 | — | 91.7 | 92.6 | — |
| 🇺🇸 Claude Mythos 5Anthropic · USA | 89.3 | 80.0 | — | 1507 | 88.0 | 91.7 | 94.1 | — |
| 🇺🇸 Claude Opus 5Anthropic · USA | 86.8 | 79.2 | 73.6 | 1493 | 84.6 | — | — | — |
| 🇺🇸 GPT-5.6 SolOpenAI · USA | 82.7 | 64.6 | 72.7 | 1483 | 91.9 | — | 94.6 | — |
| 🇺🇸 GPT-5.6 TerraOpenAI · USA | 77.9 | 63.4 | 69.6 | 1466 | 87.4 | — | 92.9 | — |
| 🇺🇸 Grok 4.5xAI · USA | 77.8 | 64.7 | 53.8 | 1471 | 83.3 | — | 93.1 | — |
| 🇨🇳 GLM-5.2Zhipu AI · China | 76.9 | 62.1 | 69.0 | 1475 | 81.0 | 80.2 | 78 | 7 |
| 🇺🇸 Claude Opus 4.8Anthropic · USA | 76.5 | 69.2 | 59.0 | 1455 | 82.7 | 88.8 | 84 | 14 |
| 🇺🇸 GPT-5.5OpenAI · USA | 76.3 | 58.6 | 67.0 | 1475 | 82.7 | — | 85 | 16 |
| 🇺🇸 Claude Sonnet 5Anthropic · USA | 74.7 | 63.2 | 53.8 | 1461 | 80.4 | — | 83 | 12 |
| 🇺🇸 Grok 4xAI · USA | 74.2 | — | 66.7 | 1430 | 83.3 | 79.4 | 84 | 16 |
| 🇺🇸 GPT-5.6 LunaOpenAI · USA | 73.7 | 62.7 | 67.2 | — | 84.7 | — | 92.3 | — |
| 🇨🇳 Qwen3.7-MaxAlibaba · China | 72.6 | 60.6 | 57.5 | 1475 | 69.7 | 91.6 | 81 | 7 |
| 🇨🇳 Kimi K2.6Moonshot AI · China | 68.4 | 58.6 | — | 1460 | 66.7 | 89.6 | 78 | 9 |
| 🇨🇳 DeepSeek V4-ProDeepSeek · China | 66.1 | 55.4 | 62.8 | 1450 | 67.9 | 93.5 | 82 | 9 |
| 🇺🇸 Gemini 3 ProGoogle · USA | 65.8 | 43.3 | 11.7 | 1501 | 54.2 | 91.7 | 84 | 15 |
| 🇺🇸 Claude Sonnet 4.6Anthropic · USA | 63.4 | 58.1 | 29.9 | 1430 | 67.0 | — | 80 | 9 |
| 🇺🇸 MAI-Thinking-1Microsoft · USA | 49.4 | 52.8 | — | — | 46.0 | 87.7 | 84.2 | — |
| 🇺🇸 Gemini 3.7 FlashGoogle · USA | — | — | 65.3 | 1491 | — | — | — | — |
| 🇺🇸 Llama 4 MaverickMeta · USA | — | — | — | 1288 | — | 43.4 | 70 | 5 |
| 🇪🇺 Mistral Large 3Mistral AI · Europa | — | — | — | 1418 | — | 74 | 72 | 6 |
| 🇪🇺 MagistralMistral AI · Europa | — | — | — | — | — | 70.88 | 70.07 | 4 |
| 🇨🇳 GLM-5.3-FlashZhipu AI · China | — | — | 63.4 | 1474 | — | — | — | — |
| 🆕 🇺🇸 GPT-6 AstraOpenAI · USA | — | — | — | — | — | — | 96.0 | — |
| 🆕 🇺🇸 Gemini 3.8 FlashGoogle · USA | — | — | 73.8 | 1494 | — | — | — | — |
| 🆕 🇺🇸 Grok 4.6xAI · USA | — | — | 66.7 | — | — | — | — | — |
Quality = equal-thirds average of SWE-bench-Pro (SW development) + LMArena (human preference) + Terminal-Bench (agentic capability), the three with reliable sources (Zendoric Quality); if one is missing its weight is shared among those present (at least two; otherwise «—»). LiveCodeBench and GPQA are shown for reference (indicative, may be incomplete) but are NOT in the index; ARC-AGI-2 (arcprize.org) tracks AGI progress: models score VERY low → still far from AGI. %, except LMArena (Elo).
💵 Economics (USD / 1M tokens)
| Model | Input | Cache | Output |
|---|---|---|---|
| 🇨🇳 Kimi K3Moonshot AI · China | $3.0 | $0.3 | $15.0 |
| 🆕 🇺🇸 Claude Fable 5.1Anthropic · USA | $10.0 | $0.25 | $50.0 |
| 🇺🇸 Claude Fable 5Anthropic · USA | $10.0 | $1.0 | $50.0 |
| 🇺🇸 Claude Mythos 5Anthropic · USA | $10.0 | $1.0 | $50.0 |
| 🇺🇸 Claude Opus 5Anthropic · USA | $5.0 | $0.5 | $25.0 |
| 🇺🇸 GPT-5.6 SolOpenAI · USA | $5.0 | $0.5 | $30.0 |
| 🇺🇸 GPT-5.6 TerraOpenAI · USA | $2.0 | $0.2 | $12.0 |
| 🇺🇸 Grok 4.5xAI · USA | $2.0 | $0.3 | $6.0 |
| 🇨🇳 GLM-5.2Zhipu AI · China | $0.6 | $0.26 | $2.2 |
| 🇺🇸 Claude Opus 4.8Anthropic · USA | $5.0 | $0.5 | $25.0 |
| 🇺🇸 GPT-5.5OpenAI · USA | $5.0 | $0.5 | $30.0 |
| 🇺🇸 Claude Sonnet 5Anthropic · USA | until Aug 31, 2026 $2.0 from Sep 1, 2026 $3.0 | until Aug 31, 2026 $0.2 from Sep 1, 2026 $0.3 | until Aug 31, 2026 $10.0 from Sep 1, 2026 $15.0 |
| 🇺🇸 Grok 4xAI · USA | $3.0 | $0.75 | $15.0 |
| 🇺🇸 GPT-5.6 LunaOpenAI · USA | $0.2 | $0.02 | $1.2 |
| 🇨🇳 Qwen3.7-MaxAlibaba · China | $1.2 | $0.25 | $6.0 |
| 🇨🇳 Kimi K2.6Moonshot AI · China | $0.6 | $0.16 | $2.5 |
| 🇨🇳 DeepSeek V4-ProDeepSeek · China | $0.28 | $0.03 | $0.87 |
| 🇺🇸 Gemini 3 ProGoogle · USA | $1.25 | $0.31 | $10.0 |
| 🇺🇸 Claude Sonnet 4.6Anthropic · USA | $3.0 | $0.3 | $15.0 |
| 🇺🇸 MAI-Thinking-1Microsoft · USA | $2.0 | — | $8.0 |
| 🇺🇸 Gemini 3.7 FlashGoogle · USA | $0.75 | $0.07 | $3.75 |
| 🇺🇸 Llama 4 MaverickMeta · USA | $0.2 | — | $0.6 |
| 🇪🇺 Mistral Large 3Mistral AI · Europa | $2.0 | $0.05 | $6.0 |
| 🇪🇺 MagistralMistral AI · Europa | $0.5 | $0.01 | $1.5 |
| 🇨🇳 GLM-5.3-FlashZhipu AI · China | $0.07 | $0.01 | $0.25 |
| 🆕 🇺🇸 GPT-6 AstraOpenAI · USA | $10.0 | $1.0 | $50.0 |
| 🆕 🇺🇸 Gemini 3.8 FlashGoogle · USA | $0.75 | — | $3.75 |
| 🆕 🇺🇸 Grok 4.6xAI · USA | $2.0 | $0.5 | $6.0 |
Claude Sonnet 5: scheduled price increase (same model) — reduced pricing until Aug 31, 2026 and standard pricing from Sep 1, 2026.
🔓 Open source & type
| Model | Open source | License | Type |
|---|---|---|---|
| 🇨🇳 Kimi K3Moonshot AI · China | Yes | Modified MIT | Open-weight |
| 🆕 🇺🇸 Claude Fable 5.1Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Claude Fable 5Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Claude Mythos 5Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Claude Opus 5Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 GPT-5.6 SolOpenAI · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 GPT-5.6 TerraOpenAI · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Grok 4.5xAI · USA | No | Proprietary | Proprietary (API only) |
| 🇨🇳 GLM-5.2Zhipu AI · China | Yes | MIT | Open-weight |
| 🇺🇸 Claude Opus 4.8Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 GPT-5.5OpenAI · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Claude Sonnet 5Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Grok 4xAI · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 GPT-5.6 LunaOpenAI · USA | No | Proprietary | Proprietary (API only) |
| 🇨🇳 Qwen3.7-MaxAlibaba · China | No | Proprietary | Proprietary (API only) |
| 🇨🇳 Kimi K2.6Moonshot AI · China | Yes | Modified MIT | Open-weight |
| 🇨🇳 DeepSeek V4-ProDeepSeek · China | Yes | MIT | Open-weight |
| 🇺🇸 Gemini 3 ProGoogle · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Claude Sonnet 4.6Anthropic · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 MAI-Thinking-1Microsoft · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Gemini 3.7 FlashGoogle · USA | No | Proprietary | Proprietary (API only) |
| 🇺🇸 Llama 4 MaverickMeta · USA | Yes | Llama 4 Community | Open-weight |
| 🇪🇺 Mistral Large 3Mistral AI · Europa | Yes | Apache-2.0 | Open-weight |
| 🇪🇺 MagistralMistral AI · Europa | Yes | Apache-2.0 | Open-weight |
| 🇨🇳 GLM-5.3-FlashZhipu AI · China | Yes | MIT | Open-weight |
| 🆕 🇺🇸 GPT-6 AstraOpenAI · USA | No | Proprietary | Proprietary (API only) |
| 🆕 🇺🇸 Gemini 3.8 FlashGoogle · USA | No | Proprietary | Proprietary (API only) |
| 🆕 🇺🇸 Grok 4.6xAI · USA | No | Proprietary | Proprietary (API only) |
🖥️ Open source you can self-host
Small/medium models you can run on your own machine (laptop/PC/Mac). Quality = Artificial Analysis Intelligence Index (0-100; output quality), the measure with best coverage of small open models (LMArena does not list sub-32B). Memory estimated at 4-bit (Q4) and 8-bit (Q8) quantization; on Apple Silicon it is UNIFIED memory (RAM=VRAM).
| Model | Quality (AA Index) | GPQA | Params | RAM Q4 | RAM Q8 | GPU | CPU / Mac | License |
|---|---|---|---|---|---|---|---|---|
| Qwen3.5 27BAlibaba | 42 | 85.5 | 27B | 16 GB | 31 GB | ≥16 GB | Limitado (mejor GPU/Mac ≥32 GB) | Apache-2.0 |
| Gemma 4 31BGoogle | 39 | 84.3 | 31B | 19 GB | 36 GB | ≥24 GB | Limitado (mejor GPU/Mac ≥32 GB) | Gemma |
| Qwen3.5 35B A3BAlibaba | 37 | 84.2 | 35B | 21 GB | 40 GB | ≥24 GB | Limitado (mejor GPU/Mac ≥32 GB) | Apache-2.0 |
| Gemma 4 26B A4BGoogle | 31 | 82.3 | 26B | 16 GB | 30 GB | ≥16 GB | Limitado (mejor GPU/Mac ≥32 GB) | Gemma |
| Nemotron-Cascade-2-30B-A3BNVIDIA | 28 | 76.1 | 30B | 18 GB | 34 GB | ≥24 GB | Limitado (mejor GPU/Mac ≥32 GB) | NVIDIA Open Model |
| gpt-oss-20bOpenAI | 24 | 71.5 | 20B | 13 GB | 25 GB | ≥16 GB | Limitado (mejor GPU/Mac ≥32 GB) | Apache-2.0 |
| Gemma 4 12BGoogle | 22 | 78.8 | 12B | 8 GB | 15 GB | ≥8 GB | Sí (CPU lento · Mac 16 GB) | Gemma |
| Gemma 4 E4BGoogle | 19 | 58.6 | 4B | 4 GB | 6 GB | ≥8 GB | Sí (CPU/Mac, fluido) | Gemma |
| Gemma 4 E2BGoogle | 15 | 43.4 | 2B | 3 GB | 4 GB | ≥8 GB | Sí (CPU/Mac, fluido) | Gemma |
🗄️ Large open source (server / multi-GPU)
Powerful open models that need a server or multiple GPUs. Quality = LMArena Elo (human preference over output, source lmarena.ai), which does cover large models. For MoE, memory counts total parameters (all experts are loaded). Memory estimated at 4-bit (Q4) and 8-bit (Q8) quantization; on Apple Silicon it is UNIFIED memory (RAM=VRAM).
| Model | Quality (LMArena) | GPQA | Params | RAM Q4 | RAM Q8 | GPU | CPU / Mac | License |
|---|---|---|---|---|---|---|---|---|
| GLM-5.2Zhipu AI | 1472 | 91.2 | 400B | 222 GB | 442 GB | 3× 80 GB (servidor) | No (servidor GPU) | MIT |
| DeepSeek-V4-Pro-0813DeepSeek | 1465 | 90.1 | 1600B | 882 GB | 1762 GB | 12× 80 GB (servidor) | No (servidor GPU) | MIT |
| Kimi K2.6Moonshot AI | 1460 | 90.5 | 1100B | 606 GB | 1212 GB | 8× 80 GB (servidor) | No (servidor GPU) | Modified MIT |
| Qwen3.5-397B-A17BAlibaba | 1450 | 88.4 | 397B | 220 GB | 438 GB | 3× 80 GB (servidor) | No (servidor GPU) | Apache-2.0 |
| Llama 4 MaverickMeta | 1420 | 69.8 | 400B | 222 GB | 442 GB | 3× 80 GB (servidor) | No (servidor GPU) | Llama 4 Community |
| Mistral Large 3Mistral AI | 1416 | 68.0 | 675B | 373 GB | 744 GB | 5× 80 GB (servidor) | No (servidor GPU) | Apache-2.0 |
| gpt-oss-120bOpenAI | 1352 | 80.1 | 117B | 66 GB | 130 GB | ≥80 GB | No (servidor GPU) | Apache-2.0 |
🎨 Image generation
The leading IMAGE generation models, split between remote services (pay per image via API) and open models you can run on your own hardware.
☁️ Remote (API / cloud)
| Model | Quality (arena) | Price (per image) | Notes |
|---|---|---|---|
| GPT Image 2OpenAI · USA | 1369 AA nº1 (era 1338); 2º MAI-Image-2.6-Prev 1350, gap 19; Reve 2.1 3º 1323. arena.ai 1382 nº1 | $0,211/img high 1024² (=$211/1K API OpenAI); fal.ai $0,005 low a $0,401 high 4K — CONFIRMADO al detalle | Nº1 t2i confirmado (AA y arena.ai). Lanzado abr-2026. Modelo con razonamiento; texto y prompt adherence líder |
| MAI-Image-2.5Microsoft · USA | 1303 AA (era 1269), nº6 t2i (era nº3); superada por MAI-Image-2.6-Preview 1350 nº2 y 2.5-Pro 1292 nº8 | Azure Foundry $47/M img out confirmado; $0,048/img confirmado (AA $48,1/1K) | Microsoft. Nueva gen MAI-Image-2.6-Preview (ago-2026) nº2 t2i y nº1 editing. Variantes Pro $108,5/1K y Flash $20/1K |
| Nano Banana Pro (Gemini 3 Pro Image)Google · USA | 1296 AA t2i (era 1222), nº7 de 155 (era nº9 de 81); superada por Nano Banana 2 (1320, nº4) | $0,134/img 1-2K · $0,24/img 4K Gemini API (Batch 50% → 4K ~$0,12) confirmado | = Gemini 3 Pro Image. Salida hasta 4096×4096. GA jun-2026, SynthID; activo, sin deprecación |
| Nano Banana 2 Lite (Gemini 3.1 Flash Lite)Google · USA | 1287 AA t2i (era 1262), nº9 de 155 (era nº4) | $0,034/img (1K) Gemini API confirmado (Batch $0,017) | = Gemini 3.1 Flash Lite Image. Rápida/barata ~4s (no verif.), solo 1K |
| GPT Image 1.5OpenAI · USA | 1304 AA (era 1260), nº5 t2i (sube, era nº6) | $0,009 low · $0,034 medio · $0,133 high (1024² OpenAI/fal.ai) — CONFIRMADO al detalle | Gen. anterior OpenAI (dic-2025), aún muy competitiva: nº5 AA, por encima de MAI-2.5. Sin deprecación |
| Grok ImaginexAI · USA | 1234 AA image-quality nº13 (era 1203/nº12); base 1216 nº24. Nuevo Imagine 2.0 nº3 arena.ai 1316 | $0,02/img base; quality $0,05 (1K)/$0,07 (2K); nuevo grok-imagine-image-2.0 $0,04/img (docs x.ai) | xAI. CORREGIDO: i2v NO es nº1; grok-imagine-video-1.5 nº5 AA (1109). Nº1 i2v es Minimax H3 Max (1203) |
| Cosmos 3 SuperNVIDIA · USA | AA t2i 1184 agentic (r46) y 1176 base; ya NO nº1 open (FLUX.2 [dev] 1197 delante). Debutó 1219 nº1 open | Open weights OpenMDW-1.1 (self-host gratis, uso comercial); API $0,04/img en fal (+$0,02 prompt expansion) | NVIDIA, omnimodal MoT dual-tower 64B desde Qwen3-VL 32B. Superado en open por FLUX.2 [dev] |
💻 Local (open-weight, on your hardware)
| Model | License | Hardware | Notes |
|---|---|---|---|
| FLUX.2Black Forest Labs | open-weight + API | GPU dedicada (variantes dev/FP8) | Lanz. 25-nov-2025; max dic-2025; klein 15-ene-2026 (4B Apache 2.0; 9B disputada). FLUX.3 ya salió (jul-2026, cerrado) |
| Qwen-Image 2.0Alibaba | Apache-2.0 | 7B — GPU 16GB+ | Lanz. 10-feb-2026, decoder 7B + enc. Qwen3-VL 8B, 2K nativo. Sucedido por Qwen-Image-3.0 (GA 5-ago-2026, cerrado) |
| Z-Image TurboAlibaba (Tongyi) | open-weight | ~1 s/imagen en H100; corre en Mac (MLX) | Alibaba Tongyi-MAI, 6B S3-DiT, 8 NFE, sub-segundo en H100. Familia: Base, Edit, Omni-Base |
| Stable Diffusion 3.5Stability AI | Stability Community | GPU 8-16GB | Vigente ago-2026, no retirado, sigue flagship de Stability. Mayor ecosistema (Civitai, ComfyUI, LoRAs) |
| HiDream-O1-Image-1.5HiDream | open-weight | GPU dedicada | v1.5 cerrada (hosted), hasta 2K, pixel-native UiT sin VAE. Dev abierto (MIT). Debutó #3 AA, hoy #16 |
Data as of 2026-09-04 · sources: Artificial Analysis Image/Video Arena · arena.ai · llm-stats.com · Pixazo · documentación de fabricantes · arena Elo is blind human preference over the output; indicative prices — verify before deciding.
🎬 Video generation
The leading VIDEO generation models, split between remote services (pay per generated second via API) and open models for your own hardware.
☁️ Remote (API / cloud)
| Model | Quality (arena) | Price (per second) | Notes |
|---|---|---|---|
| Kling 3.0 / TurboKuaishou · China | AA t2v c/audio 1108 #9 (Pro 1080p) · sin audio ~1246 · i2v 1071 #14. llm-stats 1934 #1; ausente de arena.ai | fal Turbo Pro 1080p $0,14/s; Turbo Std 720p $0,112/s. AA Kling 3.0 Pro 1080p $20,16/min (~$0,34/s) | Kuaishou 5-feb-2026 (3.0/Omni); Turbo 17-jun-2026. Hasta 15s, audio nativo, multi-shot 6 clips. Sin Kling 3.5 |
| Veo 3.1Google · USA | AA c/audio: 3.1 = 1092 #13, Lite 1089 #14, Fast 1086 #17. i2v 1086 #9. arena.ai Veo 3.1 Audio 1364 #10 | Gemini API: Lite $0,05/s 720p · Fast $0,10/s ($0,12 1080p) · Std $0,40/s 720p/1080p, $0,60/s 4K (audio incl.) | Tres niveles + audio nativo (Vertex/Gemini API). Superado por Gemini Omni Flash/1.1 en AA y arena.ai |
| Gemini Omni FlashGoogle · USA | AA t2v c/audio 1237 #2 (lidera Wan 3.0 1242); sin audio 1325 #1. i2v 1179 #4 c/audio, 1365 #1 sin audio | VERIFICADO ~$0,10/s (720p) Gemini API. Omni 1.1: $0,03/s 360p · $0,10/s 720p · $0,15/s 1080p · $0,30/s 4K | Google I/O may-2026. NUEVA VERSIÓN Gemini Omni 1.1 Flash (29-ago-2026): 40s, 4K upscale, #1 arena.ai 1515 |
| Seedance 2.0 (Dreamina)ByteDance · China | AA t2v 720p 1221 #5 c/audio · 1267 #4 sin audio. i2v 1190 #2 c/audio, 1336 #3 sin audio. arena.ai 1479 #4 | AA $9,07/min 720p (~$0,15/s). EvoLink 720p Std $0,199/s, Fast $0,161/s; 480p $0,092/$0,074. Audio incluido | ByteDance 12-feb-2026, 15s + audio. SUPERADO: Seedance 2.5 (31-jul-2026, API 7-ago): 30s, 50 refs, 4K |
| Sora 2OpenAI · USA | Ausente de AA. arena.ai t2v: Sora 2 Pro 1365 #9, Sora 2 1341 #17 (46 modelos, datos 25-ago-2026) | Sigue listado: sora-2 $0,10/s 720p; Pro $0,30/$0,50/$0,70 (720p/1024p/1080p). Batch -50% | RETIRADA: app 26-abr-2026; API 24-sep-2026 → 410 Gone. Aviso 24-mar-2026. Sin sucesor anunciado |
💻 Local (open-weight, on your hardware)
| Model | License | Hardware | Notes |
|---|---|---|---|
| Wan 2.7Alibaba | open-weight | 16-24GB VRAM (14B) | Alibaba, abr-2026 (7-abr según Alibaba Cloud). T2V/I2V/ref-to-video + edición, audio nativo, 15s. Sucesor Wan 3.0 (1242 #1) |
| LTX-2.3Lightricks | open-weight | desde 12GB VRAM (FP8: flujos de 32GB) | Lightricks, 22B DiT, audio+vídeo en una pasada; ~18x más rápido que Wan 2.2 (NO VERIF); sucesor LTX-2.5 ya publicado |
| HunyuanVideoTencent | open-weight | 16-24GB VRAM | Tencent 13B, open dic-2024. CORREGIDO: sucesor HunyuanVideo 1.5 (8.3B, ~14GB VRAM, 20-nov-2025) NO es Apache 2.0 |
| Mochi 1Genmo | open-weight | fine-tune en 1×H100/A100 80GB | Genmo, 10B AsymmDiT, oct-2024, Apache 2.0; 480p preview. Repo sin novedades desde nov-2024; sin sucesor ni cierre |
| SkyReels V1Skywork | open-weight | GPU dedicada | Skywork, 18-feb-2025, primer open human-centric (fine-tune HunyuanVideo), T2V+I2V; sucesores V2 abr-2025, V3 ene-2026, V4 feb-2026 |
Data as of 2026-09-04 · sources: Artificial Analysis Image/Video Arena · arena.ai · llm-stats.com · Pixazo · documentación de fabricantes · arena Elo is blind human preference over the output; indicative prices — verify before deciding.