Atlas

Benchmarks

TCSExam Lite

GitHub

TCSExam-Lite: 20 of TCSExam's tasks, five per track, chosen for coverage, difficulty and cost, scored with the same Index. Its 15 non-Solver tasks reuse each agent's Full submissions, and its Solver track runs self-tests locally.

Models
9
Top score
72.3
Leader
Claude Opus 5.5
Updated
Nov 5, 2026
Resolution
—
Saturation
—
Signal
92ndpercentile
205 benchmarks from 0.8 to 3.6219 benchmarks from 3.6 to 6.4134 benchmarks from 6.4 to 9.393 benchmarks from 9.3 to 12.155 benchmarks from 12.1 to 14.925 benchmarks from 14.9 to 17.713 benchmarks from 17.7 to 20.55 benchmarks from 20.5 to 23.37 benchmarks from 23.3 to 26.27 benchmarks from 26.2 to 29.01 benchmark from 29.0 to 31.83 benchmarks from 31.8 to 34.66 benchmarks from 34.6 to 37.43 benchmarks from 37.4 to 40.21 benchmark from 40.2 to 43.11 benchmark from 45.9 to 48.71 benchmark from 65.6 to 68.4

61st strongest of 779

Signal-to-noise +16.16 for the TCSExam family

Difficulty
Not estimated

Another benchmark represents this family in the fit.

9 dated models · 3 frontier modelsClosed weightsOpen weights
Data: Moonshot AIatlas.village.dev
#ModelSettingScoreSource
1Claude Opus 5.5max · Claude Code
72.3
TCSExam paper Table 3 (Lite, Opus 5.5)
2GPT-6 Astramax · Codex
67.0
TCSExam paper Table 3 (Lite, GPT-6 Astra)
3Kimi K3.1 Promax · Kimi Code
53.5
TCSExam paper Table 3 (Lite, Kimi K3.1 Pro)
4Kimi K3max · Kimi Code
48.6
TCSExam paper Table 3 (Lite, Kimi K3)
5GLM-5.3max · Claude Code
47.0
TCSExam paper Table 3 (Lite, GLM-5.3)
6Qwen3.8 Max (0902)max · Claude Code
45.4
TCSExam paper Table 3 (Lite, Qwen3.8 Max (0902))
7DeepSeek V4.1 Flash 1017max · Claude Code
42.4
TCSExam paper Table 3 (Lite, DeepSeek V4.1 Flash 1017)
8Grok 4.6x-high · Grok Build
42.1
TCSExam paper Table 3 (Lite, Grok 4.6)
9DeepSeek V4.1 Flashmax · Claude Code
39.0
TCSExam paper Table 3 (Lite, DeepSeek V4.1 Flash 0910)
Loading Atlas data…