Skip to content
VECTOR WIREAI INTELLIGENCE
UTC

Toolathlon leaderboard — the family board.

4 TRACKED VARIANTS · 2 FIRST-CLASS BOARDS

One family, 4 tracked prints: Toolathlon Verified is cited standalone and keeps live board of its own. Scoring configs and splits of the base board render as sections of this page, each under its own honest label.

● LIVE DATA48 MODELS ON THIS BOARDLEADER GLM 5.3 FLASH AT 78.4

First-class boards in this familyCITED STANDALONE · EACH KEEPS ITS OWN LIVE BOARD

#BoardModels scored
01ToolathlonTHIS PAGE48

The board — ToolathlonTOP 48 OF 48 MEASURED · BEST SCORE PER MODEL, DEFAULT VARIANT

#ModelVendorScoreVLast seen
01GLM 5.3 FlashZ.ai78.42026-08-27
02Muse Spark 1.1Meta75.62026-08-23
03Hy4Tencent74.12026-08-31
04Qwen3.8 Flash NextAlibaba73.52026-08-27
05Kimi K3Moonshot73.22026-08-23
06GLM-5.3Z.ai732026-08-23
07Qwen3.8 Max PreviewAlibaba72.52026-08-23
08DeepSeek-V4-FlashDeepSeek70.32026-08-23
09Claude Fable 5Anthropic61.72026-06-30
10Claude Mythos 5Anthropic61.12026-06-30
11Claude Opus 4.8Anthropic59.92026-08-23
12Claude Opus 4.7Anthropic59.32026-09-28
13GPT-5.6 SolOpenAI582026-08-23
14Claude Opus 4.6Anthropic56.82026-07-04
15Gemini 3.5 FlashGoogle56.52026-08-24
16GPT-5.5OpenAI55.62026-09-28
17GPT-5.4OpenAI54.62026-08-23
18Inkling-SmallThinking Machines54.42026-08-23
19Claude Sonnet 5Anthropic54.32026-08-23
20GPT-5.6 LunaOpenAI53.42026-08-23
21GPT-5.6 TerraOpenAI53.12026-08-23
22DeepSeek-V4-ProDeepSeek51.82026-08-23
23Seed 2.1 Pro PreviewByteDance50.62026-09-28
24Kimi K2.6Moonshot502026-08-23
25Laguna S 2.1poolside49.72026-08-23
26Gemini 3 Flash PreviewGoogle49.42026-08-23
27Claude Sonnet 4.6Anthropic49.42026-08-12
28Seed2.1ByteDance49.12026-09-28
29Gemini 3.1 ProGoogle48.82026-09-28
30Hy3-previewTencent48.52026-08-23
31GLM-5.2Z.ai48.22026-08-23
32GPT-5.2OpenAI46.32026-08-23
33MiniMax M2.7MiniMax46.32026-08-23
34MiniMax M2.1MiniMax43.52026-08-23
35Claude Opus 4.5Anthropic43.52026-06-05
36GPT-5.4 miniOpenAI42.92026-08-23
37Claude Sonnet 4.5Anthropic412026-06-30
38GLM-5.1Z.ai40.72026-08-23
39Qwen3.6 PlusAlibaba39.82026-08-23
40Qwen3.5 397B A17BAlibaba38.32026-08-23
41Gemini 3 ProGoogle36.42026-06-05
42GPT-5.4 nanoOpenAI35.52026-08-23
43DeepSeek-V3.2DeepSeek35.22026-09-24
44DeepSeek-V3.2-SpecialeDeepSeek35.22026-08-23
45Kimi K2.5Moonshot27.82026-05-03
46Qwen3.6 35B A3BAlibaba26.92026-08-23
47Qwen3 Max (Reasoning)Alibaba18.82026-09-09
48MiniMax M2MiniMax16.72026-06-05
◆ IV — INDEPENDENTLY VERIFIED · ◆ AA — AGGREGATOR · ◆ VA — VENDOR · ◇ SA — SOURCE-ATTRIBUTED

Variants of this boardSCORING CONFIGS & SPLITS — NOT CITED STANDALONE, SO THEY LIVE HERE

Toolathlon Pass@∞

VARIANT8 models · best score per model
#ModelVendorScoreVLast seen
01Claude Fable 5Anthropic55.62026-07-07
02Claude Mythos 5Anthropic55.62026-07-07
03Claude Opus 4.7Anthropic52.82026-07-07
04Claude Opus 4.8Anthropic48.12026-07-07
05Claude Opus 4.6Anthropic47.22026-07-07
06Claude Sonnet 5Anthropic40.72026-07-07
07Claude Sonnet 4.6Anthropic382026-07-07
08Claude Sonnet 4.5Anthropic28.72026-07-07

Toolathlon Avg turns

VARIANT7 models · best score per model
#ModelVendorScoreVLast seen
01Claude Sonnet 4.5Anthropic322026-06-30
02Claude Opus 4.7Anthropic25.92026-06-30
03Claude Opus 4.8Anthropic24.52026-06-30
04Claude Fable 5Anthropic19.82026-06-30
05Claude Mythos 5Anthropic17.62026-06-30
06Claude Opus 4.6Anthropic16.92026-06-30
07Claude Sonnet 4.6Anthropic16.52026-06-30
THE FAMILY BOARD · EVERY PRINT LABELED · EVERY SCORE SOURCE-ATTRIBUTED ON ITS OWN LIVE BOARDAll leaderboardsThe frontier boardModels by vendor