Boards

Open boards, with the licence on the score.

Image, video, vision, search and document ratings are LMArena (CC-BY-4.0). Embedding tables are MTEB (CC0-1.0). Speech stays off this page: those result files are not public without a Hugging Face token. Text leaderboards stay on Benchmarks.

Text to image

gpt-image-2.5-sunburst

1424.5 · rank 1

LMArena leaderboard dataset · CC-BY-4.0

Search

gpt-5.6-sol-xhigh

1261.1 · rank 1

LMArena leaderboard dataset · CC-BY-4.0

Code embeddings

codefuse-ai/F2LLM-v2-14B

0.807 · rank 1

MTEB results · CC0-1.0

Text to image

Checked

LMArena text-to-image board, overall category. Higher is the board's Elo-style rating. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-10-07.

Rank	Model	Score	Votes
1	gpt-image-2.5-sunburstOpenAI	1424.5	17,752
2	gpt-image-2.5-flareOpenAI	1397.6	16,889
3	gpt-image-2 (medium)OpenAI	1383.0	93,334
4	grok-imagine-image-2.0 (canvas)xAI	1333.5	3,113
5	mai-image-2.6Microsoft	1332.1	25,091
6	gemini-nano-banana-2.1Google	1327.8	5,310
7	reve-2.1Reve	1301.7	8,249
8	grok-imagine-image-2.0 (20260801)xAI	1294.2	14,483
9	muse-imageMeta	1274.0	42,262
10	reve-2.0Reve	1269.2	15,773
11	gemini-3.1-flash-image (nano-banana-2) [web-search]Google	1261.8	60,515
12	seedream-5.0-proBytedance	1255.5	94,181
13	qwen-image-3.0-proAlibaba	1255.2	11,440
14	mai-image-2.5Microsoft	1254.2	61,265
15	gemini-3.1-flash-lite-image (nano-banana-2-lite)Google	1250.2	17,804
16	gemini-3-pro-image-2k (nano-banana-pro)Google	1247.9	175,734
17	gpt-image-1.5-high-fidelityOpenAI	1237.4	169,927
18	gemini-3-pro-image-preview (nano-banana-pro)Google	1234.1	86,464
19	qwen-image-2.1Alibaba	1222.9	11,441
20	ideogram-4.0-qualityIdeogram	1205.1	60,591

Image edit

Checked

LMArena image-edit board, overall category. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-10-20.

Rank	Model	Score	Votes
1	GPT-Image-2.5 SunburstOpenAI	1514.4	82,078
2	GPT-Image-2.5 FlareOpenAI	1480.7	79,506
3	GPT-Image-2 (Medium)OpenAI	1460.6	328,307
4	Grok Imagine Image 2.0xAI	1439.4	30,542
5	Gemini Nano Banana 2.1Google	1426.2	12,938
6	MAI-Image-2.6Microsoft	1425.3	69,611
7	Grok Imagine Image 2.0 (20260801)xAI	1415.0	76,730
8	Krea 3 UltraKrea	1410.5	2,619
9	MAI-Image-2.5Microsoft	1399.9	190,119
10	Muse ImageMeta	1398.0	164,982
11	Seedream 5.0 ProBytedance	1391.3	307,763
12	Grok Imagine Image (Quality, 20260519)xAI	1390.5	38,207
13	Gemini 3 Pro Image (nano-banana-pro, 2K)Google	1390.2	676,298
14	Gemini 3.1 Flash Image (nano-banana-2, Web Search)Google	1387.0	253,535
15	ChatGPT Image Latest (High Fidelity, 20251216)OpenAI	1386.7	651,019
16	Gemini 3 Pro Image Preview (nano-banana-pro)Google	1385.2	540,728
17	Reve 2.1Reve	1373.2	20,810
18	GPT-Image-1.5 (High Fidelity)OpenAI	1367.5	676,312
19	Qwen-Image-2.1Alibaba	1362.4	36,713
20	Reve 2.0Reve	1357.8	19,270

Image to video

Checked

LMArena image-to-video board, overall category. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-09-22.

Rank	Model	Score	Votes
1	minimax-h3MiniMax	1495.4	57,112
2	gemini-omni-1.1-flashGoogle	1488.5	3,734
3	wan3.0Alibaba	1479.8	3,442
4	dreamina-seedance-2.5-720pBytedance	1477.4	15,729
5	dreamina-seedance-2.0-720pBytedance	1474.6	121,427
6	gemini-omni-flashGoogle	1464.8	87,155
7	hidream-o1-video-1.0Hidream	1456.5	6,488
8	grok-imagine-video-1.5-720pxAI	1456.0	137,459
9	flux-3-video-20260811Black Forest Labs	1449.5	37,042
10	happyhorse-1.0Aorizon	1441.7	70,640
11	wan2.7-i2vWan	1428.2	85,521
12	grok-imagine-video-720pxAI	1415.3	554,050
13	veo-3.1-audioGoogle	1398.1	25,116
14	veo-3.1-audio-1080pGoogle	1390.2	53,089
15	veo-3.1-fast-audioGoogle	1385.1	99,804
16	grok-imagine-video-480pxAI	1384.9	19,429
17	veo-3.1-fast-audio-1080pGoogle	1371.6	54,380
18	vidu-q3-proShengshu	1363.2	36,679
19	kling-v3-proKling	1354.0	206,221
20	veo-3-audioGoogle	1331.5	32,367

Text to video

Checked

LMArena text-to-video board, overall category. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-09-22.

Rank	Model	Score	Votes
1	gemini-omni-1.1-flashGoogle	1516.2	1,784
2	gemini-omni-flashGoogle	1512.9	26,576
3	flux-3-videoBlack Forest Labs	1493.0	1,302
4	grok-imagine-video-1.5-agentxAI	1492.3	1,218
5	dreamina-seedance-2.0-720pBytedance	1479.2	56,318
6	wan3.0Alibaba	1476.3	2,598
7	dreamina-seedance-2.5-720pBytedance	1473.9	8,001
8	minimax-h3MiniMax	1460.3	11,417
9	muse-videoMeta	1455.8	2,188
10	happyhorse-1.0Aorizon	1427.2	22,264
11	sora-2-proOpenAI	1367.8	54,167
12	veo-3.1-audioGoogle	1364.2	13,741
13	veo-3.1-audio-1080pGoogle	1362.2	25,076
14	veo-3.1-fast-audioGoogle	1361.9	39,484
15	veo-3.1-fast-audio-1080pGoogle	1358.0	25,870
16	veo-3-fast-audioGoogle	1347.8	25,258
17	sora-2OpenAI	1342.6	71,947
18	grok-imagine-video-720pxAI	1342.2	165,462
19	veo-3-audioGoogle	1340.2	19,002
20	wan2.7-t2vWan	1337.2	27,626

Vision

Checked

LMArena vision board, overall category. This is the multimodal board, separate from text-to-image. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-10-19.

Rank	Model	Score	Votes
1	Claude Fable 5 (High)Anthropic	1324.9	14,714
2	Claude Opus 5.5 (High)Anthropic	1320.8	3,035
3	Claude Opus 5 (High)Anthropic	1319.6	17,743
4	Claude Fable 5.1 (Max)Anthropic	1319.5	5,026
5	Claude Opus 4.7Anthropic	1316.9	23,139
6	Claude Opus 4.6 (High)Anthropic	1315.7	22,286
7	Gemini 3.8 Flash (High)Google	1315.4	4,645
8	Gemini 3.7 Flash (High)Google	1314.2	5,369
9	Claude Opus 4.7 (High)Anthropic	1313.1	22,731
10	Qwen3.8-MaxAlibaba	1312.9	9,957
11	Claude Opus 4.6Anthropic	1312.4	26,841
12	Muse Spark 1.3 (Max)Meta	1308.4	5,392
13	Gemini 3.5 Flash (High)Google	1308.3	13,862
14	Gemini 3.5 Flash (Medium)Google	1306.3	14,061
15	Muse SparkMeta	1305.8	5,865
16	Gemini 3 ProGoogle	1304.8	13,623
17	Muse Spark 1.2 (xHigh)Meta	1304.2	1,948
18	GPT-5.4 (High)OpenAI	1302.6	30,747
19	Muse Spark 1.3 (xHigh)Meta	1301.8	4,897
20	Gemini 3.6 Flash (High)Google	1298.9	8,979

Search

Checked

LMArena search board, overall category. Higher is the board's Elo-style rating. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-10-13.

Rank	Model	Score	Votes
1	gpt-5.6-sol-xhighOpenAI	1261.1	60,437
2	qwen3.8-maxAlibaba	1261.1	1,870
3	claude-opus-4-6-searchAnthropic	1251.4	173,191
4	gpt-5.5-searchOpenAI	1246.5	125,957
5	claude-fable-5.1-highAnthropic	1245.8	12,782
6	claude-opus-5-highAnthropic	1241.6	13,285
7	gpt-6-astra-maxOpenAI	1233.6	18,483
8	gpt-6.1-sol-maxOpenAI	1233.1	5,555
9	claude-opus-4-7Anthropic	1230.2	127,506
10	claude-fable-5-highAnthropic	1229.1	73,596
11	ernie-5.1Baidu	1226.7	4,024
12	claude-sonnet-4-6-searchAnthropic	1220.5	173,986
13	grok-4.5xAI	1215.9	76,458
14	claude-opus-5.5-mediumAnthropic	1212.8	10,842
15	gemini-3.1-pro-groundingGoogle	1208.8	151,436
16	gpt-5.2-searchOpenAI	1207.2	55,634
17	gemini-3-pro-groundingGoogle	1206.5	38,857
18	grok-4.20-multi-agent-beta-0309xAI	1204.8	126,238
19	claude-opus-4-8Anthropic	1204.1	106,304
20	gpt-5.1-searchOpenAI	1199.7	63,042

Documents

Checked

LMArena document board, overall category. Higher is the board's Elo-style rating. LMArena leaderboard dataset · CC-BY-4.0. Published 2026-09-13.

Rank	Model	Score	Votes
1	claude-opus-5-highAnthropic	1516.3	8,794
2	claude-fable-5.1-maxAnthropic	1512.6	1,403
3	claude-opus-4-6-highAnthropic	1507.3	27,929
4	claude-opus-4-6Anthropic	1507.0	41,260
5	claude-fable-5Anthropic	1496.1	8,497
6	claude-opus-4-7Anthropic	1495.0	22,192
7	claude-opus-4-7-highAnthropic	1494.7	21,957
8	gpt-5.5OpenAI	1485.7	21,279
9	gpt-5.5-highOpenAI	1484.5	20,944
10	gpt-5.6-sol-xhighOpenAI	1482.7	4,818
11	claude-sonnet-4-6Anthropic	1482.2	57,813
12	claude-opus-4-8-highAnthropic	1474.8	11,551
13	gpt-5.6-terra-xhighOpenAI	1471.6	5,787
14	gpt-5.4OpenAI	1471.0	33,331
15	muse-spark-1.3-maxMeta	1470.7	1,006
16	gpt-6-astra-maxOpenAI	1467.5	1,621
17	claude-sonnet-5-highAnthropic	1466.1	7,411
18	muse-spark-1.1Meta	1464.7	4,885
19	claude-opus-4-8Anthropic	1464.1	12,128
20	gemini-3.5-flash-mediumGoogle	1463.3	6,500

Embeddings

Checked

MTEB English v2. The score is the mean of task-type averages on the 0–1 scale the benchmark publishes. MTEB results · CC0-1.0.

Rank	Model	Score	Zero-shot
1	jcorners/ingot-8b-r3Jcorners	0.700	82%
2	infgrad/Jasper-Token-Compression-600MInfgrad	0.685	48%
3	Kingsoft-LLM/QZhou-EmbeddingKingsoft-LLM	0.695	48%
4	ByteDance-Seed/Seed1.5-EmbeddingByteDance-Seed	0.686	51%
5	Qwen/Qwen3-Embedding-8BQwen	0.687	95%
6	annamodels/LGAI-Embedding-PreviewAnnamodels	0.684	51%
7	Qwen/Qwen3-Embedding-4BQwen	0.681	95%
8	Bytedance/Seed1.6-embeddingBytedance	0.680	48%
9	google/gemini-embedding-001Google	0.677	95%
10	codefuse-ai/F2LLM-v2-8BCodefuse-Ai	0.669	43%
11	codefuse-ai/F2LLM-v2-14BCodefuse-Ai	0.665	43%
12	ai-sage/Giga-Embeddings-instruct-3B-0826Ai-Sage	0.664	-1%
13	ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826Ai-Sage	0.669	-1%
14	jinaai/jina-embeddings-v5-omni-smallJinaai	0.657	-1%
15	NovaSearch/jasper_en_vision_language_v1NovaSearch	0.666	51%
16	codefuse-ai/F2LLM-v2-4BCodefuse-Ai	0.667	43%
17	codefuse-ai/F2LLM-4BCodefuse-Ai	0.673	43%
18	jinaai/jina-embeddings-v5-omni-nanoJinaai	0.652	-1%
19	jinaai/jina-embeddings-v5-text-nanoJinaai	0.652	-1%
20	codefuse-ai/F2LLM-v2-1.7BCodefuse-Ai	0.664	43%

Code embeddings

Checked

MTEB Code v1. The score is the mean of task-type averages on the 0–1 scale the benchmark publishes. MTEB results · CC0-1.0.

Rank	Model	Score	Zero-shot
1	codefuse-ai/F2LLM-v2-14BCodefuse-Ai	0.807	58%
2	Bytedance/Seed1.6-embedding-1215Bytedance	0.807	91%
3	codefuse-ai/C2LLM-7BCodefuse-Ai	0.807	8%
4	codefuse-ai/F2LLM-v2-4BCodefuse-Ai	0.801	58%
5	Qwen/Qwen3-Embedding-4BQwen	0.801	100%
6	Qwen/Qwen3-Embedding-8BQwen	0.807	100%
7	codefuse-ai/F2LLM-v2-8BCodefuse-Ai	0.802	58%
8	codefuse-ai/F2LLM-v2-1.7BCodefuse-Ai	0.788	58%
9	ai-sage/Giga-Embeddings-instruct-10B-A1.8B-0826Ai-Sage	0.784	-1%
10	perplexity-ai/pplx-embed-v1-4bPerplexity	0.787	100%
11	codefuse-ai/F2LLM-v2-0.6BCodefuse-Ai	0.774	58%
12	ICT-TIME-and-Querit/ICT-TIME-and-Querit-embedding-v1ICT-TIME-and-Querit	0.780	25%
13	ai-sage/Giga-Embeddings-instruct-3B-0826Ai-Sage	0.769	-1%
14	lightonai/LateOn-CodeLightonai	0.741	8%
15	google/gemini-embedding-001Google	0.760	100%
16	lightonai/mSlideOnLightonai	0.748	0%
17	google/embeddinggemma-2Google	0.779	100%
18	naver/splade-code-8BNaver	0.789	-1%
19	codefuse-ai/F2LLM-v2-330MCodefuse-Ai	0.757	58%
20	perplexity-ai/pplx-embed-v1-0.6bPerplexity	0.758	100%