Telegram

Leaderboards

Which model is best, for what.

One ranking per job, not one ranking for everything. Rebuilt every day from data the benchmark operators publish themselves — each board names its source and the date that source last updated.

Last rebuilt

Best overall chat model

Head-to-head human preference across everyday prompts. The closest thing there is to a general ranking.

LMArena 2026-10-15

Rank# Model Made by Arena score Votes
1 Gemini 4 Argon (High) Google 1,532 4,904
2 Claude Opus 5.5 (High) Anthropic 1,516 8,300
3 Claude Fable 5.1 (Max) Anthropic 1,512 14,590
4 Claude Opus 5 (Max) Anthropic 1,508 32,495
5 Claude Opus 5 (High) Anthropic 1,503 66,824
6 Claude Opus 4.6 (High) Anthropic 1,502 81,394
7 Muse Wonder (Max) Meta 1,501 3,247
8 Gemini 3.8 Flash (High) Google 1,499 37,454
9 Claude Opus 4.6 Anthropic 1,498 86,071
10 Claude Fable 5 (High) Anthropic 1,491 42,004
11 Mimo V2.6 Pro Open Xiaomi 1,491 4,075
12 Muse Spark 1.3 (Max) Meta 1,490 17,561
13 Claude Opus 4.7 (High) Anthropic 1,490 65,116
14 Gemini 3.7 Flash (High) Google 1,487 27,232
15 Muse Spark 1.2 (xHigh) Meta 1,484 8,248
16 Claude Opus 4.7 Anthropic 1,483 66,216
17 Qwen3.8 Max Alibaba 1,483 27,766
18 Gemini 3.5 Flash High Google 1,482 48,486
19 Gemini 3.1 Pro Preview Google 1,480 127,649
20 Gemini 3 Pro Google 1,479 42,072
21 Gemini 3.6 Flash High Google 1,479 36,122
22 Muse Spark 1.1 Meta 1,478 42,810
23 Gemini 3.5 Flash Medium Google 1,477 47,292
24 Qwen3.7 Max Preview Alibaba 1,475 3,934
25 Kimi K3 (Max) Open Moonshot AI 1,475 32,795

Best model for coding

Judged on real build tasks where a human picks the better result, not on a self-reported pass rate.

LMArena 2026-10-16

Rank# Model Made by Arena score Votes
1 Claude Opus 5.5 (Max) Anthropic 1,807 2,775
2 GPT-6 Astra (Max) OpenAI 1,777 6,014
3 Claude Sonnet 5.5 (xHigh) Anthropic 1,768 3,182
4 GPT-6.1 Sol (Max) OpenAI 1,755 2,863
5 Claude Fable 5.1 (Max) Anthropic 1,741 7,503
6 Claude Sonnet 5.5 (High) Anthropic 1,711 4,914
7 Claude Opus 5 (Max) Anthropic 1,689 18,310
8 GPT-6 Sol (Max) OpenAI 1,683 4,364
9 Gemini 4 Argon (High) Google 1,678 1,775
10 Qwen3.8-Max (0902) Alibaba 1,674 13,812
11 Qwen3.8-Max Alibaba 1,672 3,440
12 Muse Spark 1.3 (Max) Meta 1,658 9,068
13 Claude Opus 5 (High) Anthropic 1,657 24,211
14 Kimi K3 (Max) Open Moonshot AI 1,655 19,220
15 Muse Wonder (Max) Meta 1,650 1,942
16 Grok 4.7 (xHigh) xAI 1,639 5,021
17 MiMo-V2.6-Flash Open Xiaomi 1,634 5,016
18 Hy4 preview Open Tencent 1,631 4,802
19 Qwen3.8-Flash-Next Open Alibaba 1,630 8,083
20 MiMo-V2.6-Pro Open Xiaomi 1,625 4,763
21 Claude Fable 5 (High) Anthropic 1,625 15,612
22 Muse Spark 1.3 (xHigh) Meta 1,622 8,436
23 GLM-5.3 (Max) Open Zai 1,621 9,542
24 DeepSeek-V4.1-Flash (Max) Open DeepSeek 1,618 3,540
25 Grok 4.6 (High) xAI 1,617 10,179

Best agent model

Long-running tasks with tool calls, where being nearly right at step three ruins step nine.

LMArena 2026-10-16

Rank# Model Made by Score % Votes
1 Claude Opus 5.5 (Max) Anthropic 18.9 10,551
2 GPT-6 Astra (Max) OpenAI 14.1 13,168
3 Claude Fable 5.1 (Max) Anthropic 14.0 16,374
4 GPT-6.1 Sol (Max) OpenAI 13.6 13,444
5 Gemini 4 Argon (High) Google 13.2 30,489
6 Claude Sonnet 5.5 (Max) Anthropic 13.0 7,560
7 GPT-6 Sol (Max) OpenAI 10.3 7,887
8 Claude Opus 5 (Max) Anthropic 8.7 23,770
9 Claude Fable 5 (High) Anthropic 8.2 42,608
10 GPT-5.6 Sol (xHigh) OpenAI 7.1 36,545
11 Muse Wonder (Max) Meta 4.9 7,690
12 Claude Sonnet 5 (High) Anthropic 4.6 33,276
13 MiMo-V2.6-Pro Open Xiaomi 4.0 27,455
14 Muse Spark 1.3 (Max) Meta 3.8 73,912
15 Gemini 3.8 Flash (High) Google 3.7 37,692
16 Grok 4.7 (xHigh) xAI 3.2 15,127
17 GPT-5.5 (xHigh) OpenAI 3.1 54,331
18 GPT-6 Luna (Max) OpenAI 3.0 163,815
19 DeepSeek-V4.1-Flash (Max) Open DeepSeek 2.7 241,716
20 Hy4 preview Open Tencent 2.7 84,828
21 MiMo-V2.6-Flash Open Xiaomi 2.6 19,620
22 GLM-5.2 (Max) Open Zai 2.3 87,460
23 Kimi K3 (Max) Open Moonshot AI 1.9 135,415
24 GLM-5.3 (Max) Open Zai 1.8 86,541
25 Qwen3.8-Max Alibaba 1.8 68,968

Best model for images and documents

Understanding what is in a picture, a chart or a scanned page.

LMArena 2026-10-09

Rank# Model Made by Arena score Votes
1 Claude Fable 5 High Anthropic 1,324 14,175
2 Claude Opus 5.5 High Anthropic 1,321 2,508
3 Claude Opus 5 High Anthropic 1,319 16,665
4 Claude Fable 5.1 Max Anthropic 1,318 4,648
5 Claude Opus 4.7 Anthropic 1,316 23,165
6 Gemini 3.7 Flash High Google 1,316 4,524
7 Claude Opus 4.6 High Anthropic 1,316 22,304
8 Gemini 3.8 Flash High Google 1,314 3,792
9 Qwen3.8 Max Alibaba 1,314 10,017
10 Claude Opus 4.6 Anthropic 1,313 26,853
11 Claude Opus 4.7 High Anthropic 1,313 22,733
12 Gemini 3.5 Flash High Google 1,310 13,103
13 Muse Spark 1.3 Max Meta 1,309 4,550
14 Gemini 3.5 Flash Medium Google 1,309 13,180
15 Muse Spark Meta 1,306 5,872
16 Gemini 3 Pro Google 1,305 13,639
17 muse-spark-1.2 (xHigh) Meta 1,305 1,954
18 GPT 5.4 High OpenAI 1,303 29,881
19 muse-spark-1.3 (xHigh) Meta 1,301 4,077
20 Gemini 3.6 Flash High Google 1,298 8,127
21 GPT 5.5 OpenAI 1,297 27,993
22 Gemini 3.1 Pro Preview Google 1,296 46,320
23 Glm 5.3 Flash Open Zai 1,296 11,157
24 Claude Opus 4.8 High Anthropic 1,294 18,341
25 Muse Spark 1.1 Meta 1,293 11,994

Best image generator

Text-to-image, ranked by people choosing between two results.

LMArena 2026-10-07

Rank# Model Made by Arena score Votes
1 GPT Image 2.5 Sunburst OpenAI 1,425 17,752
2 GPT Image 2.5 Flare OpenAI 1,398 16,889
3 gpt-image-2 (medium) OpenAI 1,383 93,334
4 grok-imagine-image-2.0 (canvas) xAI 1,334 3,113
5 Mai Image 2.6 Microsoft-Ai 1,332 25,091
6 Gemini Nano Banana 2.1 Google 1,328 5,310
7 Reve 2.1 Reve 1,302 8,249
8 grok-imagine-image-2.0 (20260801) xAI 1,294 14,483
9 Muse Image Meta 1,274 42,262
10 Reve 2.0 Reve 1,269 15,773
11 gemini-3.1-flash-image (nano-banana-2) [web-search] Google 1,262 60,515
12 Seedream 5.0 Pro ByteDance 1,255 94,181
13 Qwen Image 3.0 Pro Alibaba 1,255 11,440
14 Mai Image 2.5 Microsoft-Ai 1,254 61,265
15 gemini-3.1-flash-lite-image (nano-banana-2-lite) Google 1,250 17,804
16 gemini-3-pro-image-2k (nano-banana-pro) Google 1,248 175,734
17 GPT Image 1.5 High Fidelity OpenAI 1,237 169,927
18 gemini-3-pro-image-preview (nano-banana-pro) Google 1,234 86,464
19 Qwen Image 2.1 Open Alibaba 1,223 11,441
20 Ideogram 4.0 Quality Open Ideogram 1,205 60,591
21 Qwen Image 2.0 Pro 2026.06.22 Alibaba 1,192 13,096
22 Uni 1.1 Max Luma AI 1,189 14,072
23 Uni 1.1 Luma AI 1,184 46,699
24 Mai Image 2 Microsoft-Ai 1,184 51,475
25 Grok Imagine Image xAI 1,170 272,085

Best video generator

Text-to-video, same method: two clips, a human picks one.

LMArena 2026-09-22

Rank# Model Made by Arena score Votes
1 Gemini Omni 1.1 Flash Google 1,516 1,784
2 Gemini Omni Flash Google 1,513 26,576
3 Flux 3 Video Bfl 1,493 1,302
4 Grok Imagine Video 1.5 Agent xAI 1,492 1,218
5 Dreamina Seedance 2.0 720p ByteDance 1,479 56,318
6 Wan3.0 Alibaba 1,476 2,598
7 Dreamina Seedance 2.5 720p ByteDance 1,474 8,001
8 Minimax H3 Open MiniMax 1,460 11,417
9 Muse Video Meta 1,456 2,188
10 Happyhorse 1.0 Aorizon 1,427 22,264
11 Sora 2 Pro OpenAI 1,368 54,167
12 Veo 3.1 Audio Google 1,364 13,741
13 Veo 3.1 Audio 1080p Google 1,362 25,076
14 Veo 3.1 Fast Audio Google 1,362 39,484
15 Veo 3.1 Fast Audio 1080p Google 1,358 25,870
16 Veo 3 Fast Audio Google 1,348 25,258
17 Sora 2 OpenAI 1,343 71,947
18 Grok Imagine Video 720p xAI 1,342 165,462
19 Veo 3 Audio Google 1,340 19,002
20 Wan2.7 T2v Wan 1,337 27,626
21 Wan2.6 T2v Alibaba 1,327 54,050
22 Seedance V1.5 Pro ByteDance 1,256 76,171
23 Veo 3 Google 1,254 15,005
24 Veo 3 Fast Google 1,248 15,283
25 Wan2.5 T2v Preview Alibaba 1,245 36,378

Cheapest capable models

Live list prices across providers. Ranking here is arithmetic, not opinion.

OpenRouter 2026-10-18

Rank# Model Made by Per 1M tokens Context
1 Mistral Nemo Mistral $0.029 131K
2 Ling 3.0 Flash inclusionAI $0.032 262K
3 gpt-oss-20b OpenAI $0.036 131K
4 Granite 4.0 Micro IBM $0.041 131K
5 Nex-N2.5-Mini Nex AGI $0.044 262K
6 gpt-oss-20b (batch) OpenAI $0.046 131K
7 DeepSeek V4 Flash Latest DeepSeek $0.048 1.049M
8 Qwen3.7 Flash Qwen $0.055 1M
9 gpt-oss-120b (batch) OpenAI $0.056 131K
10 Llama 3.1 8B Instruct Meta $0.057 131K
11 Schematron V2 Turbo Inference.net $0.060 128K
12 Nova Micro 1.0 Amazon $0.061 128K
13 Gemma 3 4B Google $0.062 131K
14 Command R7B (12-2024) Cohere $0.066 128K
15 Mercury 2.5 Inception $0.068 260K
16 GPT-5 Nano (batch) OpenAI $0.069 400K
17 DeepSeek V4.1 Flash (batch) DeepSeek $0.075 1.049M
18 Laguna XS 2.1 Poolside $0.075 262K
19 Ministral 3 8B 2512 (batch) Mistral $0.075 262K
20 Gemma 3 12B Google $0.075 131K
21 GLM 5.3 Flash (batch) Z.ai $0.083 1.049M
22 DeepSeek V4.1 Flash DeepSeek $0.085 1.049M
23 GLM Flash Latest Z.ai $0.086 1.049M
24 GLM 5.3 Flash Z.ai $0.086 1.049M
25 Solar Mini 4 Upstage $0.087 524K

How to read these

Arena scores are relative, not absolute. They come from people comparing two anonymous answers and picking one. A 20-point gap near the top is often inside the margin of error — treat the top few as a tie unless the confidence intervals separate.

A benchmark is not your workload. The board tells you which model wins on average across other people's prompts. Yours are different. Use it to pick two candidates, then test both on your own work.

Prices move faster than rankings. The price board is list price per million tokens, blended three parts input to one part output. Providers discount, cache and rate-limit differently, so it is a starting point, not a bill.

We do not run these tests. Every number here is published by the operator named next to it and copied without adjustment. If a board looks stale, its source has not updated — the date shown is theirs, not ours.