GLM-4.6 vs GPT-OSS 20B
Change modelsGLM-4.6Z.AI
GPT-OSS 20BOpenAI
IndexAll categories, out of 100.
41.8 ±5.4 best
41.7 ±4.6
GapIndex points below the top.
Top
−0.1 · no clear gap
CoverageIndex weight that has results.
85%
90% best
SpeedOutput tokens per second.
30/s
310/s best
EloLMArena rating and rank.
1440 best
1288
ReleasedMonth of release.
2025-09
2025-08
Input / 1MUS dollars per 1M input tokens.
$0.50
$0.03 best
Output / 1MUS dollars per 1M output tokens.
$2.00
$0.13 best
1K chats2K tokens in, 500 out, each.
$2.00
$0.125 best
1K documents20K tokens in, 1K out, each.
$12.00
$0.73 best
1K agent tasks200K tokens in, 20K out, each.
$140.00
$8.60 best
ContextMaximum tokens in one request.
205K best
131K
AgenticMulti-step tasks with tools.
48.8 best
39.8
CodingCode writing and repair.
46.7
48.9 best
ReasoningLogic problems and puzzles.
31.1
35.3 best
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
39.0 best
36.0
MultilingualTasks in many languages.
—
37.0
InstructionTasks with strict prompt rules.
23.3
49.9 best
MathMath problems.
44.9
45.9 best
Benchmarks
15 sharedOnly GLM-4.6: 12Only GPT-OSS 20B: 22
GLM-4.6
GPT-OSS 20B