Model comparison
Codestral vs Gemini 2.5 Flash-Lite
Gemini 2.5 Flash-Lite is the stronger model overall, scoring 37.0 to 30.6 on the Noometry Index.
Last verified . 2 shared benchmarks.
Mistral AI
GoogleSummary
- They share 2 benchmarks with published results for both. Codestral scores higher in 0 categories and Gemini 2.5 Flash-Lite in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in coding, where Gemini 2.5 Flash-Lite leads 38.5 to 27.3.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 32.5% for Codestral and 40.5% for Gemini 2.5 Flash-Lite.
- Gemini 2.5 Flash-Lite is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.30 / $0.90 for Codestral.
- Gemini 2.5 Flash-Lite accepts more context: 1.05M tokens versus 256K.
Side by side
| Codestral | Gemini 2.5 Flash-Lite | |
|---|---|---|
| Provider | Mistral AI | |
| Noometry Index | 30.6 | 37.0 |
| Released | 2024-05-29 | 2025-06-17 |
| Weights | Proprietary | Proprietary |
| Context window | 256K | 1.05M |
| Max output | 8K | 66K |
| Input $ / M tokens | $0.30 | $0.10 |
| Output $ / M tokens | $0.90 | $0.40 |
| Results tracked | 7 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.5 Flash-Lite leads
Codestral: 27.3 (#321), Gemini 2.5 Flash-Lite: 38.5 (#173)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| ALE-Bench | 137.78 | 325.9 |
| Aider Polyglot | 11.1% | — |
| WeirdML | — | 35.2% |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1373 |
| BigCodeBench Complete | 52.5% | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Agentic & Tool Use Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 28.0 (#96)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 36.9% |
Reasoning Gemini 2.5 Flash-Lite leads
Codestral: 19.8 (#251), Gemini 2.5 Flash-Lite: 22.2 (#205)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | 40.5% |
| LMArena Hard Prompts | — | 1377 |
| DTBench | — | 62.8% |
| LMCA | — | 18.1% |
| Epoch Capabilities Index | — | 133.94 |
Math Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 38.0 (#144)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| Omni-MATH | — | 48% |
| LMArena Math | — | 1373 |
Knowledge Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 32.5 (#210)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| MMLU-Pro | — | 53.7% |
| Vectara Hallucination Rate | — | 3.3% |
| GPQA (HELM) | — | 30.9% |
| LMArena Expert | — | 1373 |
Multimodal Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 29.1 (#114)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| LMArena Vision | — | 1198 |
| VPCT | — | 30% |
Multilingual Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 49.3 (#134)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| LMArena Non-English | — | 1369 |
| LMArena Chinese | — | 1404 |
| LMArena French | — | 1388 |
| LMArena German | — | 1389 |
| LMArena Japanese | — | 1359 |
| LMArena Korean | — | 1360 |
| LMArena Russian | — | 1373 |
| LMArena Spanish | — | 1396 |
Instruction Following Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 70.0 (#168)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| IFEval | — | 81% |
| LMArena Instruction Following | — | 1367 |
Long Context Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 33.3 (#262)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| Fiction.LiveBench | — | 47.2% |
| LMArena Longer Query | — | 1373 |
Writing & Preference Not comparable
Codestral: —, Gemini 2.5 Flash-Lite: 56.8 (#135)
| Benchmark | Codestral | Gemini 2.5 Flash-Lite |
|---|---|---|
| LMArena Text | — | 1379 |
| LMArena Creative Writing | — | 1367 |
| WildBench | — | 81.8% |
| LMArena Multi-Turn | — | 1366 |
Frequently asked questions
Is Codestral better than Gemini 2.5 Flash-Lite?
Gemini 2.5 Flash-Lite is the stronger model overall, scoring 37.0 to 30.6 on the Noometry Index.
Which is cheaper, Codestral or Gemini 2.5 Flash-Lite?
Gemini 2.5 Flash-Lite is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Codestral lists at $0.30 and $0.90.
Is Codestral or Gemini 2.5 Flash-Lite better for coding?
Gemini 2.5 Flash-Lite scores higher on coding benchmarks: 38.5 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Gemini 2.5 Flash-Lite does, with 1.05M tokens against 256K.
How many benchmarks do Codestral and Gemini 2.5 Flash-Lite share?
2 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Gemini 2.5 Flash-Lite has 33.