Model comparison
Codestral vs Granite 4.2 8B
Granite 4.2 8B is the stronger model overall, scoring 40.5 to 30.6 on the Noometry Index.
Last verified . 0 shared benchmarks.
Mistral AISummary
- The widest gap is in coding, where Granite 4.2 8B leads 40.5 to 27.3.
- Granite 4.2 8B is cheaper at $0.06 / $0.25 per million input/output tokens, against $0.30 / $0.90 for Codestral.
- Codestral accepts more context: 256K tokens versus 131K.
- Granite 4.2 8B has downloadable open weights; the other is API-only.
Side by side
| Codestral | Granite 4.2 8B | |
|---|---|---|
| Provider | Mistral AI | IBM |
| Noometry Index | 30.6 | 40.5 |
| Released | 2024-05-29 | — |
| Weights | Proprietary | Open |
| Context window | 256K | 131K |
| Max output | 8K | 118K |
| Input $ / M tokens | $0.30 | $0.06 |
| Output $ / M tokens | $0.90 | $0.25 |
| Results tracked | 7 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Granite 4.2 8B leads
Codestral: 27.3 (#321), Granite 4.2 8B: 40.5 (#137)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| Aider Polyglot | 11.1% | — |
| BigCodeBench Instruct | 41.8% | — |
| LMArena Coding | — | 1380 |
| BigCodeBench Complete | 52.5% | — |
| ALE-Bench | 137.78 | — |
| HumanEval+ | 73.8% | — |
| MBPP+ | 61.9% | — |
Reasoning Granite 4.2 8B leads
Codestral: 19.8 (#251), Granite 4.2 8B: 26.6 (#131)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| Kagi LLM Benchmark | 32.5% | — |
| LMArena Hard Prompts | — | 1329 |
Knowledge Not comparable
Codestral: —, Granite 4.2 8B: 38.4 (#145)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| LMArena Expert | — | 1384 |
Multilingual Not comparable
Codestral: —, Granite 4.2 8B: 44.5 (#178)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| LMArena Non-English | — | 1302 |
| LMArena Chinese | — | 1366 |
| LMArena Russian | — | 1285 |
Instruction Following Not comparable
Codestral: —, Granite 4.2 8B: 68.7 (#184)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| LMArena Instruction Following | — | 1301 |
Long Context Not comparable
Codestral: —, Granite 4.2 8B: 40.3 (#159)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| LMArena Longer Query | — | 1324 |
Writing & Preference Not comparable
Codestral: —, Granite 4.2 8B: 49.6 (#189)
| Benchmark | Codestral | Granite 4.2 8B |
|---|---|---|
| LMArena Text | — | 1320 |
| LMArena Creative Writing | — | 1236 |
| LMArena Multi-Turn | — | 1301 |
Frequently asked questions
Is Codestral better than Granite 4.2 8B?
Granite 4.2 8B is the stronger model overall, scoring 40.5 to 30.6 on the Noometry Index.
Which is cheaper, Codestral or Granite 4.2 8B?
Granite 4.2 8B is cheaper. It lists at $0.06 per million input tokens and $0.25 per million output tokens; Codestral lists at $0.30 and $0.90.
Is Codestral or Granite 4.2 8B better for coding?
Granite 4.2 8B scores higher on coding benchmarks: 40.5 versus 27.3 in the Noometry coding category.
Which has the bigger context window?
Codestral does, with 256K tokens against 131K.
How many benchmarks do Codestral and Granite 4.2 8B share?
0 benchmarks have published results for both models. Codestral has 7 scored results on Noometry and Granite 4.2 8B has 11.