Kimi K2vsgpt-oss-20b

Kimi K2
gpt-oss-20b
Specifications
Parameters
1T21B
Context window
128k128k
API pricing
Cheapest input
$0.57Novita$0.03CoreWeave
Cheapest output
$2.30Novita$0.13CoreWeave
Benchmarks
SWE-Bench Verified
65.8%60.7%
GPQA Diamond
75.1%71.5%
Benchmarks
BullshitBench v2
10%
Humanity's Last Exam · no tools
10.9%
Humanity's Last Exam · with tools
17.3%
MMLU
85.3%
Overview
CompanyMoonshot AIOpenAI
Release dateJul 11 2025Aug 5 2025
AccessOpen WeightOpen Weight

Other comparisons

Kimi K2vsClaude Opus 5gpt-oss-20bvsClaude Opus 5Kimi K2vsGemini 3.7 Flashgpt-oss-20bvsGemini 3.7 FlashKimi K2vsMuse Glimmergpt-oss-20bvsMuse GlimmerKimi K2vsGrok 4.6gpt-oss-20bvsGrok 4.6Kimi K2vsDeepSeek-V4-Pro-0813gpt-oss-20bvsDeepSeek-V4-Pro-0813Kimi K2vsMistral Medium 3.5gpt-oss-20bvsMistral Medium 3.5

Frequently asked questions

Kimi K2 leads gpt-oss-20b on 2 of the 2 benchmarks they both report (SWE-Bench Verified, GPQA Diamond). Kimi K2 shipped 25 days before gpt-oss-20b, so benchmark comparisons should account for the intervening progress.

Kimi K2 has 1T parameters, while gpt-oss-20b has 21B. Context windows are 128k (Kimi K2) vs 128k (gpt-oss-20b).

On SWE-Bench Verified, Kimi K2 leads at 65.8% vs gpt-oss-20b at 60.7%. On GPQA Diamond, Kimi K2 leads at 75.1% vs gpt-oss-20b at 71.5%.

Kimi K2 was released by Moonshot AI on Jul 11 2025.

gpt-oss-20b was released by OpenAI on Aug 5 2025.

Kimi K2 leads on SWE-Bench Verified — Kimi K2 65.8% vs gpt-oss-20b 60.7%.

Kimi K2 leads on GPQA Diamond — Kimi K2 75.1% vs gpt-oss-20b 71.5%.

Kimi K2 has a 128k context window; gpt-oss-20b has 128k.