Skip to content

Policy

An Elo number is attested only when it comes from the Arena text / overall leaderboard (resource above).[^arena-text]

  • Store the leaderboard's integer score (the number before ±).
  • Map each arena name onto catalog ids using the matching policy.
  • Scores for catalog ids that are not on the board are estimated. They live in the auto-fill section of ModelCatalog.arenaScores and must not be quoted as leaderboard facts.

Replace this snapshot only via the refresh playbook.

Snapshot

Field Value
Board Text arena, overall, style control as shown on the page
Fetched 2026-08-12 (page date Aug 12, 2026)
Models on board 390
Votes (page) 7,775,261
Catalog ids receiving an attested score 328
Catalog ids still estimated 707

Attested (text arena)

Arena name → Elo → catalog ids that carry this score after the refresh.

  • 1 claude-fable-5 — 1507 (±5) → claude-fable-5

  • 2 claude-opus-4-6-thinking — 1505 (±4) → claude-opus-4-6-thinking

  • 3 claude-opus-4-7-thinking — 1502 (±4) → claude-opus-4-7-thinking

  • 4 muse-spark-1.2 (xHigh) — 1499 (±10) → muse-spark-1.2-xhigh

  • 5 claude-opus-4-6 — 1497 (±3) → claude-opus-4-6, claude-opus-4.6

  • 6 claude-opus-4-7 — 1494 (±4) → claude-opus-4-7, claude-opus-4.7

  • 7 claude-opus-5-high — 1494 (±5) → claude-opus-5-high

  • 8 claude-opus-5-max — 1491 (±7) → claude-opus-5-max

  • 9 qwen3.8-max — 1491 (±8) → qwen3.8-max

  • 10 muse-spark-1.1 — 1489 (±6) → muse-spark-1.1

  • 11 kimi-k3-max — 1489 (±6) → kimi-k3-max

  • 12 muse-spark — 1488 (±6) → muse-spark

  • 13 gemini-3.1-pro-preview — 1486 (±3) → gemini-3.1-pro-preview

  • 14 gemini-3-pro — 1485 (±4) → gemini-3-pro

  • 15 gemini-3.6-flash — 1484 (±6) → gemini-3.6-flash

  • 16 claude-opus-4-8-thinking — 1482 (±5) → claude-opus-4-8-thinking

  • 17 gpt-5.5-high — 1482 (±4) → gpt-5.5-high

  • 18 gpt-5.6-sol-xhigh — 1481 (±6) → gpt-5.6-sol-xhigh

  • 19 gemini-3.5-flash-high — 1477 (±5) → gemini-3.5-flash-high

  • 20 gpt-5.5 — 1477 (±4) → gpt-5.5

  • 21 gpt-5.4-high — 1477 (±4) → gpt-5.4-high

  • 22 gpt-5.2-chat-latest-20260210 — 1476 (±4) → gpt-5.2-chat-latest-20260210

  • 23 gemini-3.5-flash-medium — 1475 (±5) → gemini-3.5-flash-medium

  • 24 grok-4.20-beta1 — 1475 (±5) → grok-4.20-beta1

  • 25 qwen3.7-max-preview — 1474 (±10) → qwen3.7-max-preview

  • 26 claude-opus-4-8 — 1474 (±5) → claude-opus-4-8, claude-opus-4.8

  • 27 gpt-5.5-instant — 1473 (±5) → gpt-5.5-instant

  • 28 gemini-3-flash — 1473 (±4) → gemini-3-flash

  • 29 claude-opus-4-5-20251101-thinking-32k — 1473 (±4) → claude-opus-4-5-20251101-thinking-32k

  • 30 claude-sonnet-4-6 — 1472 (±4) → claude-sonnet-4-6, claude-sonnet-4.6

  • 31 grok-4.20-beta-0309-reasoning — 1472 (±4) → grok-4.20-beta-0309-reasoning

  • 32 glm-5.2-max — 1471 (±5) → glm-5.2-max

  • 33 grok-4.20-multi-agent-beta-0309 — 1471 (±4) → grok-4.20-multi-agent-beta-0309

  • 34 claude-opus-4-5-20251101 — 1469 (±3) → claude-opus-4-5-20251101

  • 35 grok-4.5 — 1468 (±5) → grok-4.5

  • 36 ernie-5.1 — 1468 (±5) → ernie-5.1

  • 37 mimo-v2.5-pro — 1468 (±4) → mimo-v2.5-pro

  • 38 glm-5.1 — 1467 (±4) → glm-5.1

  • 39 qwen3.5-max-preview — 1466 (±5) → qwen3.5-max-preview

  • 40 grok-4.1-thinking — 1465 (±3) → grok-4.1-thinking

  • 41 gpt-5.4 — 1465 (±4) → gpt-5.4

  • 42 gpt-5.6-terra-xhigh — 1465 (±6) → gpt-5.6-terra-xhigh

  • 43 grok-4.6-high — 1464 (±12) → grok-4-6-high, grok-4.6-high

  • 44 claude-sonnet-5-high — 1462 (±5) → claude-sonnet-5-high

  • 45 kimi-k2.6 — 1461 (±5) → kimi-k2-6, kimi-k2.6

  • 46 qwen3.6-max-preview — 1460 (±8) → qwen3.6-max-preview

  • 47 grok-4.1 — 1459 (±3) → grok-4.1

  • 48 qwen3.7-plus — 1458 (±5) → qwen3.7-plus

  • 49 gemini-3-flash (thinking-minimal) — 1458 (±3) → gemini-3-flash-thinking-minimal

  • 50 deepseek-v4-pro — 1458 (±4) → deepseek-v4-pro, deepseek-v4-pro:free

  • 51 gemini-3.5-flash-lite — 1458 (±6) → gemini-3.5-flash-lite

  • 52 glm-5 — 1457 (±4) → glm-5

  • 53 hy3 — 1457 (±9) → hy3

  • 54 claude-sonnet-4-5-20250929-thinking-32k — 1456 (±3) → claude-sonnet-4-5-20250929-thinking-32k

  • 55 deepseek-v4-pro-high-preview — 1456 (±4) → deepseek-v4-pro-high-preview

  • 56 dola-seed-2.0-pro — 1456 (±3) → dola-seed-2.0-pro

  • 57 claude-sonnet-4-5-20250929 — 1455 (±3) → claude-sonnet-4-5-20250929

  • 58 gpt-5.1-high — 1455 (±4) → gpt-5.1-high

  • 59 gemma-4-31b — 1451 (±8) → gemma-4-31b

  • 60 gpt-5.6-luna-xhigh — 1451 (±6) → gpt-5.6-luna-xhigh

  • 61 kimi-k2.5-thinking — 1450 (±3) → kimi-k2.5-thinking

  • 62 ernie-5.0-preview-1203 — 1449 (±7) → ernie-5.0-preview-1203

  • 63 claude-opus-4-1-20250805-thinking-16k — 1449 (±3) → claude-opus-4-1-20250805-thinking-16k

  • 64 gpt-5.3-chat-latest — 1449 (±4) → gpt-5.3-chat-latest

  • 65 mimo-v2-pro — 1448 (±5) → mimo-v2-pro

  • 66 gpt-5.4-mini-high — 1448 (±4) → gpt-5.4-mini-high

  • 67 claude-opus-4-1-20250805 — 1447 (±3) → claude-opus-4-1-20250805

  • 68 ernie-5.0-0110 — 1446 (±4) → ernie-5.0-0110

  • 69 gemini-2.5-pro — 1445 (±2) → gemini-2.5-pro

  • 70 gpt-4.5-preview-2025-02-27 — 1445 (±6) → gpt-4.5-preview-2025-02-27

  • 71 qwen3.6-plus — 1443 (±4) → qwen3.6-plus, qwen3.6-plus-free

  • 72 minimax-m3 — 1443 (±5) → minimax-m3

  • 73 chatgpt-4o-latest-20250326 — 1443 (±3) → chatgpt-4o-latest-20250326

  • 74 qwen3.5-397b-a17b — 1442 (±4) → qwen3.5-397b-a17b

  • 75 inkling — 1442 (±6) → inkling

  • 76 glm-4.7 — 1442 (±6) → glm-4.7

  • 77 grok-4.3 — 1441 (±4) → grok-4.3

  • 78 gpt-5.1 — 1439 (±4) → gpt-5.1

  • 79 deepseek-v4-flash-high-preview — 1438 (±4) → deepseek-v4-flash-high-preview

  • 80 gemma-4-26b-a4b — 1438 (±8) → gemma-4-26b-a4b

  • 81 gpt-5.2-high — 1437 (±4) → gpt-5.2-high

  • 82 longcat-flash-chat-2602-exp — 1436 (±5) → longcat-flash-chat-2602-exp

  • 83 gpt-5.2 — 1436 (±3) → gpt-5.2

  • 84 deepseek-v4-flash — 1435 (±4) → deepseek-v4-flash, deepseek-v4-flash-free, deepseek-v4-flash:free

  • 85 qwen3-max-preview — 1434 (±4) → qwen3-max-preview

  • 86 gpt-5-high — 1434 (±4) → gpt-5-high

  • 87 mimo-v2.5 — 1434 (±4) → mimo-v2.5, mimo-v2.5-free

  • 88 glm-5v-turbo — 1434 (±7) → glm-5v-turbo

  • 89 gemini-3.1-flash-lite-preview — 1432 (±4) → gemini-3.1-flash-lite-preview

  • 90 kimi-k2.5-instant — 1431 (±7) → kimi-k2.5-instant

  • 91 grok-4-1-fast-reasoning — 1431 (±3) → grok-4-1-fast-reasoning

  • 92 o3-2025-04-16 — 1431 (±4) → o3-2025-04-16

  • — AutoEval — 1431 (±11) → —

  • 93 mimo-v2-omni — 1430 (±6) → mimo-v2-omni

  • 94 kimi-k2-thinking-turbo — 1430 (±3) → kimi-k2-thinking-turbo

  • 95 mistral-medium-3.5 — 1427 (±7) → mistral-medium-3-5, mistral-medium-3.5

  • 96 gpt-5-chat — 1427 (±4) → gpt-5-chat

  • 97 nvidia-nemotron-3-ultra-550b-a55b-nvfp4 — 1427 (±7) → nvidia-nemotron-3-ultra-550b-a55b-nvfp4

  • 98 muse-glimmer — 1426 (±10) → muse-glimmer

  • 99 amazon-nova-experimental-chat-26-02-10 — 1426 (±10) → amazon-nova-experimental-chat-26-02-10

  • 100 deepseek-v3.2 — 1425 (±4) → deepseek-v3-2, deepseek-v3.2

  • 101 deepseek-v3.2-exp-thinking — 1425 (±7) → deepseek-v3.2-exp-thinking

  • 102 claude-opus-4-20250514-thinking-16k — 1425 (±4) → claude-opus-4-20250514-thinking-16k

  • 103 glm-4.6 — 1424 (±4) → glm-4.6

  • 104 qwen3-max-2025-09-23 — 1424 (±6) → qwen3-max-2025-09-23

  • 105 qwen3-235b-a22b-instruct-2507 — 1423 (±3) → qwen3-235b-a22b-instruct-2507

  • 106 deepseek-v3.2-thinking — 1423 (±4) → —

  • 107 deepseek-v3.2-exp — 1422 (±6) → deepseek-v3.2-exp

  • 108 deepseek-r1-0528 — 1422 (±6) → deepseek-r1-0528

  • 109 grok-4-fast-chat — 1420 (±8) → grok-4-fast-chat

  • 110 ernie-5.0-preview-1022 — 1418 (±9) → ernie-5.0-preview-1022

  • 111 kimi-k2-0711-preview — 1418 (±5) → kimi-k2-0711-preview

  • 112 kimi-k2-0905-preview — 1418 (±6) → kimi-k2-0905-preview

  • 113 deepseek-v3.1-terminus-thinking — 1417 (±10) → deepseek-v3.1-terminus-thinking

  • 114 deepseek-v3.1 — 1417 (±6) → deepseek-v3-1, deepseek-v3.1

  • 115 qwen3.5-122b-a10b — 1417 (±4) → qwen3.5-122b-a10b

  • 116 deepseek-v3.1-thinking — 1416 (±7) → deepseek-v3.1-thinking

  • 117 minimax-m2.7 — 1416 (±4) → minimax-m2.7

  • 118 amazon-nova-experimental-chat-26-01-10 — 1415 (±10) → amazon-nova-experimental-chat-26-01-10

  • 119 mistral-large-3 — 1415 (±3) → mistral-large-3

  • 120 deepseek-v3.1-terminus — 1415 (±10) → deepseek-v3.1-terminus

  • 121 qwen3-vl-235b-a22b-instruct — 1414 (±6) → qwen3-vl-235b-a22b-instruct

  • 122 gpt-4.1-2025-04-14 — 1414 (±4) → gpt-4.1-2025-04-14

  • 123 hunyuan-hy3-preview — 1413 (±8) → hunyuan-hy3-preview

  • 124 claude-haiku-4-5-20251001 — 1413 (±3) → claude-haiku-4-5-20251001

  • 125 claude-opus-4-20250514 — 1413 (±4) → claude-opus-4-20250514

  • 126 grok-3-preview-02-24 — 1411 (±4) → grok-3-preview-02-24

  • 127 glm-4.5 — 1411 (±5) → glm-4.5

  • 128 gemini-2.5-flash — 1410 (±2) → gemini-2.5-flash

  • 129 grok-4-0709 — 1409 (±4) → grok-4-0709

  • 130 mistral-medium-2508 — 1409 (±3) → mistral-medium-2508

  • 131 qwen3.5-27b — 1408 (±4) → qwen3.5-27b

  • 132 gemini-2.5-flash-preview-09-2025 — 1404 (±4) → gemini-2.5-flash-preview-09-2025

  • 133 grok-4-fast-reasoning — 1404 (±5) → grok-4-fast-reasoning

  • 134 qwen3-235b-a22b-no-thinking — 1403 (±5) → qwen3-235b-a22b-no-thinking

  • 135 gpt-5.4-nano-high — 1402 (±4) → gpt-5.4-nano-high

  • 136 o1-2024-12-17 — 1402 (±4) → o1-2024-12-17

  • 137 longcat-flash-chat — 1401 (±6) → longcat-flash-chat

  • 138 qwen3-next-80b-a3b-instruct — 1401 (±5) → qwen3-next-80b-a3b-instruct, qwen3-next-80b-a3b-instruct:free

  • 139 claude-sonnet-4-20250514-thinking-32k — 1400 (±4) → claude-sonnet-4-20250514-thinking-32k

  • 140 qwen3-235b-a22b-thinking-2507 — 1399 (±7) → qwen3-235b-a22b-thinking-2507

  • 141 deepseek-r1 — 1398 (±5) → deepseek-r1

  • 142 qwen3.5-flash — 1397 (±4) → qwen3.5-flash

  • 143 deepseek-v3-0324 — 1396 (±4) → deepseek-v3-0324

  • 144 qwen3.5-35b-a3b — 1395 (±4) → qwen3.5-35b-a3b

  • 145 qwen3-vl-235b-a22b-thinking — 1395 (±7) → qwen3-vl-235b-a22b-thinking

  • 146 hunyuan-vision-1.5-thinking — 1395 (±12) → hunyuan-vision-1.5-thinking

  • 147 step-3.5-flash — 1394 (±4) → step-3.5-flash

  • 148 amazon-nova-experimental-chat-12-10 — 1394 (±10) → amazon-nova-experimental-chat-12-10

  • 149 mimo-v2-flash (non-thinking) — 1392 (±4) → —

  • 150 minimax-m2.5 — 1390 (±4) → minimax-m2.5, minimax-m2.5-free, minimax-m2.5:free

  • 151 o4-mini-2025-04-16 — 1390 (±4) → o4-mini-2025-04-16

  • 152 gpt-5-mini-high — 1390 (±5) → gpt-5-mini-high

  • 153 claude-sonnet-4-20250514 — 1390 (±4) → claude-sonnet-4-20250514

  • 154 o1-preview — 1389 (±5) → o1-preview

  • 155 claude-3-7-sonnet-20250219-thinking-32k — 1388 (±4) → claude-3-7-sonnet-20250219-thinking-32k

  • 156 qwen3-coder-480b-a35b-instruct — 1388 (±5) → qwen3-coder-480b-a35b-instruct

  • 157 mistral-medium-2505 — 1387 (±5) → mistral-medium-2505

  • 158 hunyuan-t1-20250711 — 1387 (±9) → hunyuan-t1-20250711

  • 159 mimo-v2-flash (thinking) — 1386 (±6) → mimo-v2-flash-thinking

  • 160 minimax-m2.1-preview — 1384 (±5) → minimax-m2.1-preview

  • 161 qwen3-30b-a3b-instruct-2507 — 1383 (±5) → qwen3-30b-a3b-instruct-2507

  • 162 gpt-4.1-mini-2025-04-14 — 1383 (±4) → gpt-4.1-mini-2025-04-14

  • 163 hunyuan-turbos-20250416 — 1383 (±6) → hunyuan-turbos-20250416

  • 164 gemini-2.5-flash-lite-preview-09-2025-no-thinking — 1380 (±3) → gemini-2.5-flash-lite-preview-09-2025-no-thinking

  • 165 trinity-large-preview — 1379 (±4) → trinity-large-preview, trinity-large-preview:free

  • 166 solar-pro4 — 1378 (±12) → solar-pro4

  • 167 glm-4.6v — 1377 (±11) → glm-4.6v

  • 168 gemini-2.5-flash-lite-preview-06-17-thinking — 1375 (±5) → gemini-2.5-flash-lite-preview-06-17-thinking

  • 169 qwen3-235b-a22b — 1375 (±5) → qwen3-235b-a22b

  • 170 qwen2.5-max — 1374 (±4) → qwen2.5-max

  • 171 claude-3-5-sonnet-20241022 — 1373 (±3) → claude-3-5-sonnet-20241022

  • 172 glm-4.5-air — 1373 (±4) → glm-4.5-air, glm-4.5-air:free

  • 173 claude-3-7-sonnet-20250219 — 1372 (±4) → claude-3-7-sonnet-20250219

  • 174 trinity-large-thinking — 1369 (±5) → trinity-large-thinking, trinity-large-thinking:free

  • 175 qwen3-next-80b-a3b-thinking — 1369 (±6) → qwen3-next-80b-a3b-thinking

  • 176 glm-4.7-flash — 1367 (±6) → glm-4.7-flash

  • 177 gemma-3-27b-it — 1366 (±4) → gemma-3-27b-it, gemma-3-27b-it:free

  • 178 amazon-nova-experimental-chat-11-10 — 1366 (±4) → amazon-nova-experimental-chat-11-10

  • 179 o3-mini-high — 1364 (±5) → o3-mini-high

  • 180 minimax-m1 — 1363 (±4) → minimax-m1

  • 181 grok-3-mini-high — 1362 (±5) → grok-3-mini-high

  • 182 gemini-2.0-flash-001 — 1360 (±4) → gemini-2.0-flash-001

  • 183 nvidia-nemotron-3-super-120b-a12b — 1360 (±7) → nvidia-nemotron-3-super-120b-a12b

  • 184 deepseek-v3 — 1359 (±5) → deepseek-v3

  • 185 mistral-small-2506 — 1357 (±5) → mistral-small-2506

  • 186 grok-3-mini-beta — 1356 (±5) → grok-3-mini-beta

  • 187 intellect-3 — 1356 (±8) → intellect-3

  • 188 command-a-03-2025 — 1354 (±3) → command-a-03-2025

  • 189 gemini-2.0-flash-lite-preview-02-05 — 1354 (±4) → gemini-2.0-flash-lite-preview-02-05

  • 190 glm-4.5v — 1353 (±8) → glm-4.5v

  • 191 gpt-oss-120b — 1352 (±4) → gpt-oss-120b, gpt-oss-120b:free

  • 192 gemini-1.5-pro-002 — 1351 (±3) → gemini-1.5-pro-002

  • 193 nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4 — 1350 (±11) → nvidia-nemotron-3.5-lightning-30b-a3b-nvfp4

  • 194 amazon-nova-experimental-chat-10-20 — 1350 (±6) → amazon-nova-experimental-chat-10-20

  • 195 hunyuan-turbos-20250226 — 1349 (±12) → hunyuan-turbos-20250226

  • 196 step-3 — 1348 (±7) → step-3

  • 197 o3-mini — 1348 (±4) → o3-mini

  • 198 llama-3.1-nemotron-ultra-253b-v1 — 1348 (±12) → llama-3.1-nemotron-ultra-253b-v1

  • 199 amazon-nova-experimental-chat-10-09 — 1347 (±11) → amazon-nova-experimental-chat-10-09

  • 200 qwen3-32b — 1347 (±9) → qwen3-32b

  • 201 mercury-2 — 1346 (±11) → mercury-2

  • 202 qwen-plus-0125 — 1346 (±8) → qwen-plus-0125

  • 203 ling-flash-2.0 — 1346 (±7) → ling-flash-2.0

  • 204 gpt-4o-2024-05-13 — 1346 (±3) → gpt-4o-2024-05-13

  • 205 minimax-m2 — 1346 (±8) → minimax-m2

  • 206 nvidia-llama-3.3-nemotron-super-49b-v1.5 — 1343 (±10) → nvidia-llama-3.3-nemotron-super-49b-v1.5

  • 207 glm-4-plus-0111 — 1343 (±8) → glm-4-plus-0111

  • 208 claude-3-5-sonnet-20240620 — 1343 (±3) → claude-3-5-sonnet-20240620

  • 209 gemma-3-12b-it — 1342 (±10) → gemma-3-12b-it, gemma-3-12b-it:free

  • 210 hunyuan-turbo-0110 — 1341 (±12) → hunyuan-turbo-0110

  • 211 gpt-5-nano-high — 1337 (±7) → gpt-5-nano-high

  • 212 o1-mini — 1337 (±4) → o1-mini

  • 213 qwq-32b — 1336 (±4) → qwq-32b

  • 214 nova-2-lite — 1336 (±6) → nova-2-lite

  • 215 grok-2-2024-08-13 — 1336 (±4) → grok-2-2024-08-13

  • 216 gemini-advanced-0514 — 1336 (±5) → gemini-advanced-0514

  • 217 gpt-4o-2024-08-06 — 1335 (±4) → gpt-4o-2024-08-06

  • 218 llama-3.1-405b-instruct-bf16 — 1335 (±4) → —

  • 219 step-2-16k-exp-202412 — 1334 (±9) → step-2-16k-exp-202412

  • 220 llama-3.1-405b-instruct-fp8 — 1333 (±4) → —

  • 221 olmo-3.1-32b-instruct — 1329 (±6) → olmo-3.1-32b-instruct

  • 222 molmo-2-8b — 1329 (±21) → molmo-2-8b

  • 223 yi-lightning — 1328 (±5) → yi-lightning

  • 224 llama-3.3-nemotron-49b-super-v1 — 1328 (±12) → —

  • 225 qwen3-30b-a3b — 1327 (±5) → qwen3-30b-a3b

  • 226 llama-4-maverick-17b-128e-instruct — 1327 (±4) → llama-4-maverick-17b-128e-instruct

  • 227 hunyuan-large-2025-02-10 — 1326 (±10) → —

  • 228 gpt-4-turbo-2024-04-09 — 1324 (±4) → gpt-4-turbo-2024-04-09

  • 229 claude-3-5-haiku-20241022 — 1324 (±3) → claude-3-5-haiku-20241022

  • 230 gemini-1.5-pro-001 — 1324 (±4) → gemini-1.5-pro-001

  • 231 deepseek-v2.5-1210 — 1323 (±8) → deepseek-v2.5-1210

  • 232 llama-4-scout-17b-16e-instruct — 1322 (±5) → llama-4-scout-17b-16e-instruct

  • 233 gpt-4.1-nano-2025-04-14 — 1322 (±8) → gpt-4.1-nano-2025-04-14

  • 234 claude-3-opus-20240229 — 1322 (±3) → claude-3-opus-20240229

  • 235 ring-flash-2.0 — 1320 (±7) → ring-flash-2.0

  • 236 step-1o-turbo-202506 — 1320 (±7) → —

  • 237 glm-4-plus — 1319 (±5) → glm-4-plus

  • 238 llama-3.3-70b-instruct — 1318 (±3) → llama-3.3-70b-instruct, llama-3.3-70b-instruct:free

  • 239 gemma-3n-e4b-it — 1318 (±5) → gemma-3n-e4b-it, gemma-3n-e4b-it:free

  • 240 qwen-max-0919 — 1318 (±6) → —

  • 241 gpt-4o-mini-2024-07-18 — 1318 (±4) → gpt-4o-mini-2024-07-18

  • 242 gpt-oss-20b — 1318 (±6) → gpt-oss-20b, gpt-oss-20b:free

  • 243 nvidia-nemotron-3-nano-30b-a3b-bf16 — 1315 (±6) → —

  • 244 qwen2.5-plus-1127 — 1315 (±6) → —

  • 245 athene-v2-chat — 1314 (±5) → —

  • 246 mistral-large-2407 — 1314 (±4) → mistral-large-2407

  • 247 gpt-4-0125-preview — 1313 (±4) → gpt-4-0125-preview

  • 248 gpt-4-1106-preview — 1312 (±4) → gpt-4-1106-preview

  • 249 hunyuan-standard-2025-02-10 — 1311 (±10) → —

  • 250 gemini-1.5-flash-002 — 1309 (±4) → gemini-1.5-flash-002

  • 251 grok-2-mini-2024-08-13 — 1308 (±4) → —

  • 252 deepseek-v2.5 — 1307 (±5) → deepseek-v2.5

  • 253 athene-70b-0725 — 1307 (±6) → —

  • 254 mercury — 1306 (±14) → mercury

  • 255 granite-4.1-8b — 1306 (±10) → granite-4.1-8b

  • 256 mistral-large-2411 — 1305 (±4) → mistral-large-2411

  • 257 olmo-3-32b-think — 1305 (±8) → olmo-3-32b-think

  • 258 magistral-medium-2506 — 1304 (±6) → magistral-medium-2506

  • 259 mistral-small-3.1-24b-instruct-2503 — 1303 (±5) → mistral-small-3.1-24b-instruct-2503

  • 260 gemma-3-4b-it — 1303 (±9) → gemma-3-4b-it, gemma-3-4b-it:free

  • 261 qwen2.5-72b-instruct — 1303 (±4) → qwen2.5-72b-instruct

  • 262 llama-3.1-nemotron-70b-instruct — 1299 (±8) → llama-3.1-nemotron-70b-instruct

  • 263 hunyuan-large-vision — 1294 (±9) → —

  • 264 llama-3.1-70b-instruct — 1293 (±4) → llama-3.1-70b-instruct

  • 265 amazon-nova-pro-v1.0 — 1290 (±5) → —

  • 266 jamba-1.5-large — 1289 (±7) → jamba-1.5-large

  • 267 gemma-2-27b-it — 1289 (±3) → gemma-2-27b-it

  • 268 reka-core-20240904 — 1288 (±7) → reka-core-20240904

  • 269 gpt-4-0314 — 1287 (±5) → gpt-4-0314

  • 270 ibm-granite-h-small — 1287 (±8) → —

  • 271 gemini-1.5-flash-001 — 1286 (±5) → gemini-1.5-flash-001

  • 272 llama-3.1-nemotron-51b-instruct — 1286 (±10) → llama-3.1-nemotron-51b-instruct

  • 273 llama-3.1-tulu-3-70b — 1286 (±10) → —

  • 274 olmo-3.1-32b-think — 1285 (±7) → —

  • 275 claude-3-sonnet-20240229 — 1281 (±4) → claude-3-sonnet-20240229

  • 276 gemma-2-9b-it-simpo — 1280 (±7) → —

  • 277 nemotron-4-340b-instruct — 1277 (±5) → nemotron-4-340b-instruct

  • 278 llama-3-70b-instruct — 1276 (±4) → llama-3-70b-instruct

  • 279 command-r-plus-08-2024 — 1276 (±7) → command-r-plus-08-2024

  • 280 gpt-4-0613 — 1275 (±4) → gpt-4-0613

  • 281 mistral-small-24b-instruct-2501 — 1274 (±6) → mistral-small-24b-instruct-2501

  • 282 glm-4-0520 — 1273 (±7) → —

  • 283 reka-flash-20240904 — 1272 (±7) → —

  • 284 qwen2.5-coder-32b-instruct — 1270 (±8) → qwen2.5-coder-32b-instruct

  • 285 c4ai-aya-expanse-32b — 1267 (±5) → c4ai-aya-expanse-32b

  • 286 gemma-2-9b-it — 1267 (±4) → gemma-2-9b-it

  • 287 deepseek-coder-v2 — 1265 (±6) → deepseek-coder-v2

  • 288 qwen2-72b-instruct — 1261 (±5) → qwen2-72b-instruct

  • 289 command-r-plus — 1261 (±4) → command-r-plus

  • 290 claude-3-haiku-20240307 — 1261 (±4) → claude-3-haiku-20240307

  • 291 amazon-nova-lite-v1.0 — 1260 (±5) → —

  • 292 gemini-1.5-flash-8b-001 — 1259 (±4) → gemini-1.5-flash-8b-001

  • 293 phi-4 — 1256 (±5) → phi-4

  • 294 olmo-2-0325-32b-instruct — 1251 (±11) → olmo-2-0325-32b-instruct

  • 295 command-r-08-2024 — 1250 (±7) → command-r-08-2024

  • 296 mistral-large-2402 — 1242 (±5) → —

  • 297 amazon-nova-micro-v1.0 — 1241 (±5) → —

  • 298 jamba-1.5-mini — 1240 (±7) → jamba-1.5-mini

  • 299 ministral-8b-2410 — 1237 (±9) → ministral-8b-2410

  • 300 gemini-pro-dev-api — 1236 (±7) → —

  • 301 qwen1.5-110b-chat — 1234 (±6) → —

  • 302 hunyuan-standard-256k — 1233 (±12) → —

  • 303 reka-flash-21b-20240226-online — 1233 (±7) → —

  • 304 qwen1.5-72b-chat — 1233 (±5) → —

  • 305 mixtral-8x22b-instruct-v0.1 — 1229 (±5) → mixtral-8x22b-instruct-v0.1

  • 306 reka-flash-21b-20240226 — 1226 (±6) → —

  • 307 command-r — 1226 (±5) → command-r

  • 308 gpt-3.5-turbo-0125 — 1225 (±5) → gpt-3.5-turbo-0125

  • 309 llama-3-8b-instruct — 1223 (±4) → llama-3-8b-instruct

  • 310 c4ai-aya-expanse-8b — 1223 (±7) → c4ai-aya-expanse-8b

  • 311 gemini-pro — 1223 (±12) → —

  • 312 mistral-medium — 1222 (±5) → mistral-medium

  • 313 llama-3.1-tulu-3-8b — 1220 (±11) → —

  • 314 yi-1.5-34b-chat — 1213 (±5) → —

  • 315 zephyr-orpo-141b-A35b-v0.1 — 1212 (±11) → —

  • 316 llama-3.1-8b-instruct — 1211 (±4) → llama-3.1-8b-instruct

  • 317 granite-3.1-8b-instruct — 1208 (±11) → granite-3.1-8b-instruct

  • 318 qwen1.5-32b-chat — 1203 (±6) → —

  • 319 gpt-3.5-turbo-1106 — 1203 (±9) → gpt-3.5-turbo-1106

  • 320 gemma-2-2b-it — 1200 (±4) → gemma-2-2b-it

  • 321 phi-3-medium-4k-instruct — 1197 (±5) → phi-3-medium-4k-instruct

  • 322 mixtral-8x7b-instruct-v0.1 — 1197 (±4) → mixtral-8x7b-instruct-v0.1

  • 323 dbrx-instruct-preview — 1195 (±6) → —

  • 324 internlm2\_5-20b-chat — 1191 (±7) → —

  • 325 qwen1.5-14b-chat — 1191 (±7) → —

  • 326 deepseek-llm-67b-chat — 1184 (±11) → —

  • 327 wizardlm-70b — 1184 (±9) → —

  • 328 yi-34b-chat — 1183 (±7) → yi-34b-chat

  • 329 granite-3.0-8b-instruct — 1182 (±9) → granite-3.0-8b-instruct

  • 330 openchat-3.5 — 1182 (±10) → —

  • 331 openchat-3.5-0106 — 1182 (±8) → —

  • 332 gemma-1.1-7b-it — 1182 (±6) → —

  • 333 snowflake-arctic-instruct — 1179 (±6) → —

  • 334 granite-3.1-2b-instruct — 1179 (±11) → granite-3.1-2b-instruct

  • 335 tulu-2-dpo-70b — 1177 (±10) → —

  • 336 openhermes-2.5-mistral-7b — 1175 (±10) → —

  • 337 vicuna-33b — 1172 (±6) → —

  • 338 starling-lm-7b-beta — 1171 (±7) → —

  • 339 phi-3-small-8k-instruct — 1171 (±6) → phi-3-small-8k-instruct

  • 340 llama-2-70b-chat — 1170 (±5) → —

  • 341 starling-lm-7b-alpha — 1167 (±8) → —

  • 342 llama-3.2-3b-instruct — 1166 (±8) → llama-3.2-3b-instruct, llama-3.2-3b-instruct:free

  • 343 nous-hermes-2-mixtral-8x7b-dpo — 1164 (±12) → nous-hermes-2-mixtral-8x7b-dpo

  • 344 granite-3.0-2b-instruct — 1156 (±8) → granite-3.0-2b-instruct

  • 345 qwq-32b-preview — 1154 (±11) → qwq-32b-preview

  • 346 llama2-70b-steerlm-chat — 1154 (±13) → —

  • 347 solar-10.7b-instruct-v1.0 — 1152 (±13) → —

  • 348 dolphin-2.2.1-mistral-7b — 1152 (±15) → —

  • 349 mpt-30b-chat — 1150 (±12) → —

  • 350 wizardlm-13b — 1149 (±9) → —

  • 351 mistral-7b-instruct-v0.2 — 1149 (±7) → mistral-7b-instruct-v0.2

  • 352 falcon-180b-chat — 1147 (±17) → —

  • 353 qwen1.5-7b-chat — 1143 (±10) → —

  • 354 phi-3-mini-4k-instruct-june-2024 — 1143 (±6) → —

  • 355 llama-2-13b-chat — 1141 (±7) → —

  • 356 vicuna-13b — 1141 (±7) → —

  • 357 qwen-14b-chat — 1139 (±11) → —

  • 358 palm-2 — 1138 (±9) → —

  • 359 gemma-7b-it — 1137 (±9) → —

  • 360 codellama-34b-instruct — 1136 (±9) → —

  • 361 zephyr-7b-beta — 1130 (±9) → —

  • 362 phi-3-mini-128k-instruct — 1129 (±7) → phi-3-mini-128k-instruct

  • 363 phi-3-mini-4k-instruct — 1128 (±6) → phi-3-mini-4k-instruct

  • 364 guanaco-33b — 1127 (±12) → —

  • 365 zephyr-7b-alpha — 1126 (±16) → —

  • 366 stripedhyena-nous-7b — 1121 (±11) → —

  • 367 codellama-70b-instruct — 1119 (±18) → —

  • 368 gemma-1.1-2b-it — 1116 (±8) → —

  • 369 vicuna-7b — 1115 (±9) → —

  • 370 smollm2-1.7b-instruct — 1114 (±14) → —

  • 371 llama-3.2-1b-instruct — 1111 (±8) → llama-3.2-1b-instruct

  • 372 mistral-7b-instruct — 1110 (±9) → mistral-7b-instruct

  • 373 llama-2-7b-chat — 1108 (±7) → —

  • 374 gemma-2b-it — 1093 (±11) → —

  • 375 qwen1.5-4b-chat — 1090 (±9) → —

  • 376 olmo-7b-instruct — 1073 (±11) → —

  • 377 koala-13b — 1070 (±10) → —

  • 378 alpaca-13b — 1069 (±11) → —

  • 379 gpt4all-13b-snoozy — 1067 (±15) → —

  • 380 mpt-7b-chat — 1063 (±12) → —

  • 381 chatglm3-6b — 1056 (±12) → chatglm3-6b

  • 382 RWKV-4-Raven-14B — 1042 (±11) → —

  • 383 chatglm2-6b — 1024 (±14) → —

  • 384 oasst-pythia-12b — 1023 (±11) → —

  • 385 chatglm-6b — 995 (±13) → —

  • 386 fastchat-t5-3b — 992 (±12) → —

  • 387 dolly-v2-12b — 981 (±13) → —

  • 388 llama-13b — 974 (±16) → —

  • 389 stablelm-tuned-alpha-7b — 953 (±13) → —

Estimated

707 catalog ids keep a family / sibling / alias estimate (or an older attested value that this refresh did not rematch). They are not re-derived on this pass.

Do not treat those numbers as Arena facts. See matching-policy.md.

Notes

  • Runtime lookup strips a leading provider/ prefix and lowercases the id; catalog keys are already lowercase.
  • New board models added to the runtime catalog on this refresh: grok-4.6 / grok-4.6-high, muse-glimmer, solar-pro4, nemotron-3.5-lightning.
  • Historical unmatched board rows (older chat models Kai does not ship metadata for) are listed above with → — and were not added to baseEntries.

[^arena-text]: Arena text leaderboard