GPT-5.2 Codex vs Ministral 14B vs Command A Translate
Ministral 14B comes out ahead, 64 to 42 and 17 on our weighted score, and it is the cheaper option too.
OpenAIGPT-5.2 Codex
42/100- ECI—
- Price$1.75 / $14.00
- Context400K
- Our pick
Mistral AIMinistral 14B
64/100- ECI—
- Price$0.20 / $0.20
- Context262K
CohereCommand A Translate
17/100- ECI—
- Price$2.50 / $10.00
- Context8K
Ministral 14B is our pick
Ministral 14B is the better all-round choice, scoring 64/100 against GPT-5.2 Codex (42) and Command A Translate (17). It leads on price. GPT-5.2 Codex wins on inputs & features and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceMinistral 14BMinistral 14B $0.20 · Command A Translate $4.38 · GPT-5.2 Codex $4.81 per 1M tokens (3:1 blend)
- Longest contextGPT-5.2 CodexGPT-5.2 Codex 400,000 · Ministral 14B 262,144 · Command A Translate 8,000 tokens
- Widest inputsGPT-5.2 CodexGPT-5.2 Codex: Text, Images, PDFs · Ministral 14B: Text, Images · Command A Translate: Text
- Self-hostingMinistral 14B and Command A TranslatePublishes downloadable weights (Apache-2.0)
| Measure | Weight | GPT-5.2 Codex | Ministral 14B | Command A Translate |
|---|---|---|---|---|
| Price | 50% | 18 | 83 | 19 |
| Inputs & features | 30% | 80 | 50 | 25 |
| Context window | 20% | 44 | 37 | 0 |
| Overall | 100% | 42/100 | 64/100 | 17/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | GPT-5.2 CodexOpenAI | Ministral 14BMistral AI | Command A TranslateCohere |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $1.75 | $0.20 (best) | $2.50 |
| Output | $14.00 | $0.20 (best) | $10.00 |
| Cached input | — | — | — |
| Blended (3:1) | $4.81 | $0.20 (best) | $4.38 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 11 providers | Median of 1 providers | Official Cohere API |
| Limits | |||
| Context window | 400,000 tokens (best) | 262,144 tokens | 8,000 tokens |
| Max output | 128,000 tokens | 262,144 tokens (best) | 8,000 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | Yes | Yes | No |
| PDFs | Yes | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | Yes | No | No |
| Availability | |||
| Weights | Proprietary | OpenApache-2.0 | Open |
| API model ID | — | — | command-a-translate-08-2025 |
| API providers | 11 (best) | 1 | 1 |
| Released | Dec 11, 2025 | Dec 2, 2025 | Aug 28, 2025 |
| Knowledge cutoff | Aug 31, 2025 | — | Jun 1, 2024 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- GPT-5.2 Codex$45.50
- Ministral 14B$2.40
- Command A Translate$45.00
Which should you choose?
Which is better: GPT-5.2 Codex, Ministral 14B or Command A Translate?
Ministral 14B is the better all-round choice, scoring 64/100 against GPT-5.2 Codex (42) and Command A Translate (17). It leads on price. GPT-5.2 Codex wins on inputs & features and context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, GPT-5.2 Codex, Ministral 14B or Command A Translate?
Ministral 14B is cheaper at $0.20 input / $0.20 output per million tokens (median across 1 API provider). Command A Translate costs $2.50 input / $10.00 output per million tokens (official Cohere API price); GPT-5.2 Codex costs $1.75 input / $14.00 output per million tokens (median across 11 API providers). At a typical mix of three input tokens to one output token, that is $0.20 per million tokens for Ministral 14B versus $4.38 for Command A Translate (22× as much) and $4.81 for GPT-5.2 Codex (24× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. GPT-5.2 Codex has not been scored yet, Ministral 14B has not been scored yet and Command A Translate has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for GPT-5.2 Codex, Ministral 14B and Command A Translate yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
GPT-5.2 Codex has the largest context window at 400,000 tokens, against 262,144 for Ministral 14B and 8,000 for Command A Translate. Maximum output per response: GPT-5.2 Codex up to 128,000, Ministral 14B up to 262,144, Command A Translate up to 8,000 tokens.
Which can read images, PDFs, audio or video?
GPT-5.2 Codex accepts text, images and PDFs; Ministral 14B accepts text and images; Command A Translate accepts text. GPT-5.2 Codex handles the widest range of inputs.
Are any of these open source?
Ministral 14B and Command A Translate publishes its weights (Apache-2.0) and can be self-hosted; GPT-5.2 Codex is proprietary.
Which is newer?
GPT-5.2 Codex is the newest, released Dec 11, 2025. Ministral 14B came out Dec 2, 2025; Command A Translate came out Aug 28, 2025. Knowledge cutoff: GPT-5.2 Codex Aug 31, 2025, Command A Translate Jun 1, 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
