★VALS INDEX
43.4%
+/- 11.9
Accuracy
⏱VALS INDEX
10.5s
Latency (median)
$VALS INDEX
—
Cost / Test (1P-est)
SCORE STATUSscored
TEST SETS11
RECORDED TRIALS199
CUSTOM PASS67%
Industry benchmarks
Accuracyattempts · responded
GPQA_DIAMOND
GPQA Diamond
51.8%
110 attempts · 105 responded
MMLU_PRO
MMLU-Pro
35.0%
80 attempts · 66 responded
Custom 9-task suite
Mean /5n=trials
CUSTOM
agentic_tool_use
★
5.00
1 trials
CUSTOM
code_gen_long
★
5.00
1 trials
CUSTOM
json_strict
★
5.00
1 trials
CUSTOM
reasoning_multistep
★
5.00
1 trials
CUSTOM
code_debug
-
4.00
1 trials
CUSTOM
summarize
-
4.00
1 trials
CUSTOM
agentic_prompt
-
3.00
1 trials
CUSTOM
refactor_existing_code
-
3.00
1 trials
CUSTOM
creative_write
-
0.00
1 trials
Per-task distribution
Score
All trials on this page were run through the harness. Industry suites use deterministic family-specific reference graders (no LLM judge). The custom suite is graded by consensus of minimax-m3 + claude-fable-5. See /industry/ for the full benchmark suite.