Model comparison
GPT-5.2 Codex vs Kimi K2 (Jul 2025)
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 4.8× less per token, which makes it the better buy when GPT-5.2 Codex's lead doesn't matter for your workload.
Last verified . 3 shared benchmarks.
Summary
- They share 3 benchmarks with published results for both. GPT-5.2 Codex scores higher in 2 categories and Kimi K2 (Jul 2025) in 0 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.2 Codex leads 41.0 to 32.4.
- The biggest single-benchmark swing is Terminal-Bench: 66.5% for GPT-5.2 Codex and 35.7% for Kimi K2 (Jul 2025).
- Kimi K2 (Jul 2025) is cheaper at $0.57 / $2.30 per million input/output tokens, against $1.75 / $14 for GPT-5.2 Codex.
- GPT-5.2 Codex accepts more context: 400K tokens versus 262K.
- Kimi K2 (Jul 2025) has downloadable open weights; the other is API-only.
Side by side
| GPT-5.2 Codex | Kimi K2 (Jul 2025) | |
|---|---|---|
| Provider | OpenAI | Moonshot AI |
| Noometry Index | 42.6 | 41.2 |
| Released | 2025-12-18 | 2025-07-12 |
| Weights | Proprietary | Open |
| Context window | 400K | 262K |
| Max output | 128K | 262K |
| Input $ / M tokens | $1.75 | $0.57 |
| Output $ / M tokens | $14 | $2.30 |
| Results tracked | 5 | 42 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 Codex leads
GPT-5.2 Codex: 45.5 (#71), Kimi K2 (Jul 2025): 42.4 (#102)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| SWE-bench Verified (bash only) | 72.8% | 63.4% |
| ALE-Bench | 1,300 | 597.5 |
| Aider Polyglot | — | 59.1% |
| LMArena WebDev | 1339 | — |
| SWE-bench Multilingual | 66.3% | — |
| GSO | — | 4.9% |
| WeirdML | — | 42.8% |
| LMArena Coding | — | 1399 |
Agentic & Tool Use GPT-5.2 Codex leads
GPT-5.2 Codex: 41.0 (#22), Kimi K2 (Jul 2025): 32.4 (#64)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| Terminal-Bench | 66.5% | 35.7% |
| Berkeley Function Calling Leaderboard | — | 59.1% |
| METR Time Horizons | — | 59.2% |
Reasoning Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 23.3 (#179)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| SimpleBench | — | 26.3% |
| Kagi LLM Benchmark | — | 64.4% |
| LMArena Hard Prompts | — | 1384 |
| Epoch Capabilities Index | — | 146.01 |
| ForecastBench | — | 60.2 |
Math Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 42.7 (#83)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| Omni-MATH | — | 65.4% |
| LMArena Math | — | 1397 |
| FrontierMath (Feb 2025 set) | — | 21.4% |
| FrontierMath Tier 4 (v1) | — | 0% |
Knowledge Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 37.3 (#157)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| MMLU-Pro | — | 81.9% |
| Confabulations | — | 20.4% |
| Vectara Hallucination Rate | — | 17.9% |
| GPQA (HELM) | — | 65.3% |
| LMArena Expert | — | 1365 |
Multilingual Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 49.6 (#130)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Non-English | — | 1372 |
| LMArena Chinese | — | 1415 |
| LMArena French | — | 1379 |
| LMArena German | — | 1387 |
| LMArena Japanese | — | 1349 |
| LMArena Korean | — | 1325 |
| LMArena Russian | — | 1385 |
| LMArena Spanish | — | 1386 |
Instruction Following Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 71.1 (#156)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| IFEval | — | 85% |
| LMArena Instruction Following | — | 1348 |
Long Context Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 41.2 (#145)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| Fiction.LiveBench | — | 66.7% |
| CL-bench | — | 17.6% |
| LMArena Longer Query | — | 1353 |
Writing & Preference Not comparable
GPT-5.2 Codex: —, Kimi K2 (Jul 2025): 62.3 (#78)
| Benchmark | GPT-5.2 Codex | Kimi K2 (Jul 2025) |
|---|---|---|
| LMArena Text | — | 1380 |
| LMArena Creative Writing | — | 1350 |
| Short-Story Creative Writing | — | 85.6% |
| EQ-Bench Creative Writing | — | 1666 |
| WildBench | — | 86.2% |
| LMArena Multi-Turn | — | 1371 |
Frequently asked questions
Is GPT-5.2 Codex better than Kimi K2 (Jul 2025)?
GPT-5.2 Codex is the stronger model overall, scoring 42.6 to 41.2 on the Noometry Index. Kimi K2 (Jul 2025) costs 4.8× less per token, which makes it the better buy when GPT-5.2 Codex's lead doesn't matter for your workload.
Which is cheaper, GPT-5.2 Codex or Kimi K2 (Jul 2025)?
Kimi K2 (Jul 2025) is cheaper. It lists at $0.57 per million input tokens and $2.30 per million output tokens; GPT-5.2 Codex lists at $1.75 and $14.
Is GPT-5.2 Codex or Kimi K2 (Jul 2025) better for coding?
GPT-5.2 Codex scores higher on coding benchmarks: 45.5 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
GPT-5.2 Codex does, with 400K tokens against 262K.
How many benchmarks do GPT-5.2 Codex and Kimi K2 (Jul 2025) share?
3 benchmarks have published results for both models. GPT-5.2 Codex has 5 scored results on Noometry and Kimi K2 (Jul 2025) has 42.