GPT-5.3-Codex vs Mistral Large 3 675B Instruct
GPT-5.3-Codex (2026) and Mistral Large 3 675B Instruct (2025) are agentic coding models from OpenAI and MistralAI. GPT-5.3-Codex ships a not-yet-sourced context window, while Mistral Large 3 675B Instruct ships a 128K-token context window. On τ-bench, GPT-5.3-Codex leads by 7.6 pts. On pricing, Mistral Large 3 675B Instruct costs $0.5/1M input tokens versus $1.75/1M for the alternative. This comparison covers specs, pricing, capabilities, benchmarks, provider availability, and production fit.
Mistral Large 3 675B Instruct is ~250% cheaper at $0.5/1M; pay for GPT-5.3-Codex only for coding workflow support.
Specs
| Released | 2026-02-05 | 2025-12-01 |
| Context window | — | 128K |
| Parameters | — | 675B |
| Architecture | decoder only | decoder only |
| License | Proprietary | 1 |
| Knowledge cutoff | - | - |
Pricing and availability
| GPT-5.3-Codex | Mistral Large 3 675B Instruct | |
|---|---|---|
| Input price | $1.75/1M tokens | $0.5/1M tokens |
| Output price | $14/1M tokens | $1.5/1M tokens |
| Providers |
Capabilities
| GPT-5.3-Codex | Mistral Large 3 675B Instruct | |
|---|---|---|
| Vision | ||
| Multimodal | ||
| Reasoning | ||
| Function calling | ||
| Tool use | ||
| Structured outputs | ||
| Code execution |
Benchmarks
| Benchmark | GPT-5.3-Codex | Mistral Large 3 675B Instruct |
|---|---|---|
| τ-bench | 77.8 | 70.2 |
Deep dive
On shared benchmark coverage, τ-bench has GPT-5.3-Codex at 77.8 and Mistral Large 3 675B Instruct at 70.2, with GPT-5.3-Codex ahead by 7.6 points. The largest visible gap is 7.6 points on τ-bench, which matters most when that benchmark mirrors your workload. Treat isolated benchmark wins as directional, because provider routing, prompt style, and tool access can move real application results.
The capability footprint differs most on multimodal input: GPT-5.3-Codex, function calling: GPT-5.3-Codex, tool use: GPT-5.3-Codex, and code execution: GPT-5.3-Codex. Both models share structured outputs, so the practical split is not just feature count. Use those differences to decide whether the page is about raw model quality, agentic coding support, multimodal ingestion, or predictable structured API behavior.
For cost, GPT-5.3-Codex lists $1.75/1M input and $14/1M output tokens, while Mistral Large 3 675B Instruct lists $0.5/1M input and $1.5/1M output tokens on the cheapest tracked provider. A 70/30 input-output blend puts Mistral Large 3 675B Instruct lower by about $4.63 per million blended tokens. Availability is 1 providers versus 3, so concentration risk also matters.
Choose GPT-5.3-Codex when coding workflow support are central to the workload. Choose Mistral Large 3 675B Instruct when provider fit, lower input-token cost, and broader provider choice are more important. For production, rerun your own prompts through the exact provider, region, and tool stack you plan to ship.
FAQ
Which is cheaper, GPT-5.3-Codex or Mistral Large 3 675B Instruct?
Mistral Large 3 675B Instruct is cheaper on tracked token pricing. GPT-5.3-Codex costs $1.75/1M input and $14/1M output tokens. Mistral Large 3 675B Instruct costs $0.5/1M input and $1.5/1M output tokens. Provider discounts or batch pricing can still change the final bill.
Is GPT-5.3-Codex or Mistral Large 3 675B Instruct open source?
GPT-5.3-Codex is listed under Proprietary. Mistral Large 3 675B Instruct is listed under 1. License labels affect whether you can self-host, redistribute weights, or rely only on hosted APIs, so confirm the upstream license before deployment.
Which is better for multimodal input, GPT-5.3-Codex or Mistral Large 3 675B Instruct?
GPT-5.3-Codex has the clearer documented multimodal input signal in this comparison. If multimodal input is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Which is better for function calling, GPT-5.3-Codex or Mistral Large 3 675B Instruct?
GPT-5.3-Codex has the clearer documented function calling signal in this comparison. If function calling is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Which is better for tool use, GPT-5.3-Codex or Mistral Large 3 675B Instruct?
GPT-5.3-Codex has the clearer documented tool use signal in this comparison. If tool use is mission-critical, validate it against the provider endpoint because model-level support and API-level exposure can differ.
Where can I run GPT-5.3-Codex and Mistral Large 3 675B Instruct?
GPT-5.3-Codex is available on OpenRouter. Mistral Large 3 675B Instruct is available on AWS Bedrock, NVIDIA NIM, and Mistral AI Studio. Provider coverage can affect latency, region availability, compliance posture, and fallback options.
Continue comparing
Last reviewed: 2026-04-24. Data sourced from public model cards and provider documentation.