GPT-5.2-Codex vs Qwen3.8-27B
Change modelsGPT-5.2-CodexOpenAI
Qwen3.8-27BAlibaba
IndexAll categories, out of 100.
59.5 ±4.1 best
59.1 ±2.1
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
44/s
113/s best
EloLMArena rating and rank.
—
1436
ReleasedMonth of release.
2025-12
2026-08
Input / 1MUS dollars per 1M input tokens.
$1.75
$0.15 best
Output / 1MUS dollars per 1M output tokens.
$14.00
$1.88 best
1K chats2K tokens in, 500 out, each.
$10.50
$1.24 best
1K documents20K tokens in, 1K out, each.
$49.00
$4.88 best
1K agent tasks200K tokens in, 20K out, each.
$630.00
$67.50 best
ContextMaximum tokens in one request.
400K
1.05M best
AgenticMulti-step tasks with tools.
57.9 best
57.3
CodingCode writing and repair.
60.0
60.2 best
ReasoningLogic problems and puzzles.
59.9
61.3 best
MultimodalTasks with images and text.
54.4
63.4 best
KnowledgeFacts and expert knowledge.
59.9 best
51.1
MultilingualTasks in many languages.
—
60.5
InstructionTasks with strict prompt rules.
61.2
68.3 best
MathMath problems.
63.4 best
56.2
Benchmarks
17 sharedOnly GPT-5.2-Codex: 6Only Qwen3.8-27B: 83
GPT-5.2-Codex
Qwen3.8-27B