Model comparison
Codestral vs Kimi K2.5
Kimi K2.5 is the stronger model overall, scoring 48.1 to 30.6 on the Noometry Index. Codestral costs 2.0× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Mistral AISummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and Kimi K2.5 in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Kimi K2.5 leads 48.8 to 27.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 78.5% for Kimi K2.5.
- Codestral is cheaper at $0.30 / $0.90 per million input/output tokens, against $0.45 / $2.25 for Kimi K2.5.
- Kimi K2.5 accepts more context: 262K tokens versus 256K.
- Kimi K2.5 has downloadable open weights; the other is API-only.
Side by side
| Codestral | Kimi K2.5 | |
|---|---|---|
| Provider | Mistral AI | Moonshot AI |
| Noometry Index | 30.6 | 48.1 |
| Released | 2024-05-29 | 2026-01-27 |
| Weights | Proprietary | Open |
| Context window | 256K | 262K |
| Max output | 8K | 262K |
| Input $ / M tokens | $0.30 | $0.45 |
| Output $ / M tokens | $0.90 | $2.25 |
| Results tracked | 7 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Kimi K2.5 leads
Codestral: 27.3 (#321), Kimi K2.5: 48.8 (#53)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| ALE-Bench | 137.78 | 821.65 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 70.8% |
| Aider Polyglot | 11.1% | — |
| LMArena WebDev | — | 1437 |
| SWE-bench Multilingual | — | 67.3% |
| SciCode | — | 49% |
| WeirdML | — | 45.6% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1474 |
| BigCodeBench Complete | 52.5% | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, Kimi K2.5: 34.2 (#48)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| Terminal-Bench | — | 43.2% |
| OSWorld | — | 63.3% |
| Vending-Bench 2 | — | 1,198 |
Reasoning Kimi K2.5 leads
Codestral: 19.8 (#251), Kimi K2.5: 31.2 (#80)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | 78.5% |
| ARC-AGI-2 | — | 11.8% |
| SimpleBench | — | 46.8% |
| NYT Connections (extended) | — | 69.9% |
| ARC-AGI-1 | — | 65.3% |
| CritPt | — | 3.1% |
| Chess Puzzles | — | 12% |
| EnigmaEval | — | 3.4% |
| Thematic Generalization | — | 69.4% |
| LMArena Hard Prompts | — | 1453 |
| Epoch Capabilities Index | — | 148.03 |
Math Not comparable
Codestral: —, Kimi K2.5: 51.8 (#53)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| MathArena Final-Answer Competitions | — | 62.3% |
| OTIS Mock AIME 2024-2025 | — | 92.2% |
| LMArena Math | — | 1470 |
| FrontierMath (Feb 2025 set) | — | 27.9% |
| FrontierMath Tier 4 (v1) | — | 4.2% |
Knowledge Not comparable
Codestral: —, Kimi K2.5: 53.6 (#56)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| GPQA Diamond | — | 87.6% |
| Humanity's Last Exam | — | 24.4% |
| SimpleQA Verified | — | 34.3% |
| Vectara Hallucination Rate | — | 14.2% |
| LMArena Expert | — | 1466 |
Multimodal Not comparable
Codestral: —, Kimi K2.5: 41.1 (#39)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| LMArena Vision | — | 1269 |
| LMArena Document | — | 1430 |
Multilingual Not comparable
Codestral: —, Kimi K2.5: 53.9 (#53)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| LMArena Non-English | — | 1433 |
| LMArena Chinese | — | 1495 |
| LMArena French | — | 1454 |
| LMArena German | — | 1441 |
| LMArena Japanese | — | 1421 |
| LMArena Korean | — | 1410 |
| LMArena Russian | — | 1435 |
| LMArena Spanish | — | 1450 |
Instruction Following Not comparable
Codestral: —, Kimi K2.5: 75.3 (#64)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| LMArena Instruction Following | — | 1431 |
Long Context Not comparable
Codestral: —, Kimi K2.5: 52.1 (#7)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| Fiction.LiveBench | — | 86.1% |
| CL-bench | — | 19.3% |
| CL-bench Life | — | 13.2% |
| LMArena Longer Query | — | 1445 |
Writing & Preference Not comparable
Codestral: —, Kimi K2.5: 65.1 (#53)
| Benchmark | Codestral | Kimi K2.5 |
|---|---|---|
| LMArena Text | — | 1445 |
| LMArena Creative Writing | — | 1423 |
| EQ-Bench Creative Writing | — | 1579 |
| LMArena Multi-Turn | — | 1444 |
Frequently asked questions
Is Codestral better than Kimi K2.5?
Kimi K2.5 is the stronger model overall, scoring 48.1 to 30.6 on the Noometry Index. Codestral costs 2.0× less per token, which makes it the better buy when Kimi K2.5's lead doesn't matter for your workload.
Which is cheaper, Codestral or Kimi K2.5?
Codestral is cheaper. It lists at $0.30 per million input tokens and $0.90 per million output tokens; Kimi K2.5 lists at $0.45 and $2.25.
Is Codestral or Kimi K2.5 better for coding?
Kimi K2.5 scores higher on coding benchmarks: 48.8 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.5 does, with 262K tokens against 256K.
How many benchmarks do Codestral and Kimi K2.5 share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Kimi K2.5 has 51.