xai-direct · reasoning
XAI-DIRECT

grok-4.20-reasoning-direct

Grok 4.20 reasoning (2M ctx, xAI direct OAuth)

⚖ COMPARE MODELS
VALS INDEX
90.3% +/- 1.7
Accuracy
VALS INDEX
20.8s
Latency (median)
$VALS INDEX
$0.3654
Cost / Test (1P-est)
SCORE STATUSscored
TEST SETS11
RECORDED TRIALS255
CUSTOM PASS74%

Industry benchmarks

Accuracyattempts · responded
GPQA_DIAMOND GPQA Diamond
89.1%
55 attempts · 54 responded
IFEVAL IFEval
91.5%
55 attempts · 55 responded
Industry benchmark Bars = accuracy %, black cap = that score

Custom 9-task suite

Mean /5n=trials
CUSTOM json_strict
5.00
10 trials
CUSTOM reasoning_multistep
5.00
10 trials
CUSTOM code_gen_long
4.85
20 trials
CUSTOM code_debug
4.75
20 trials
CUSTOM summarize
4.60
20 trials
CUSTOM agentic_prompt
4.45
20 trials
CUSTOM agentic_tool_use
4.40
10 trials
CUSTOM refactor_existing_code -
2.15
20 trials
CUSTOM creative_write
1.00
15 trials
Custom suite · scored 0–5 by consensus (m3 + Fable-5) ★ = task scored a 5 at least once · sorted by mean score

Cross-judge validation

m3 vs Fable-5 Δ

No judge bias detected on this model — m3 and Fable-5 agree across all tasks.

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.