GPT-5 Nano vs Gemma 4 31B IT

It is close. Our weighted score puts them within a point (GPT-5 Nano 70/100, Gemma 4 31B IT 69/100), so choose by what matters most for your work: Gemma 4 31B IT for raw capability, GPT-5 Nano on price and GPT-5 Nano for long inputs. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.

GPT-5 Nano is cheaper at $0.05 input / $0.40 output per million tokens (official OpenAI API price). Gemma 4 31B IT costs $0.14 input / $0.40 output per million tokens (median across 30 API providers). At a typical mix of three input tokens to one output token, that is $0.138 per million tokens for GPT-5 Nano versus $0.205 for Gemma 4 31B IT (1.5× as much).

Gemma 4 31B IT scores higher on the Capabilities Index (ECI): Gemma 4 31B IT 142.8 (#86 of 148) and GPT-5 Nano 139.4 (#102 of 148). The confidence ranges of the top two overlap (140.3–144.9 vs 134.9–141.7), so treat the gap as small. On individual benchmarks: GPQA Diamond — Gemma 4 31B IT 75.8%, GPT-5 Nano 69.4%; OTIS Mock AIME 2024–2025 — GPT-5 Nano 81.1%, Gemma 4 31B IT 73.3%; SimpleQA Verified — GPT-5 Nano 11.7%, Gemma 4 31B IT 10.4%.

There are no published SWE-bench Verified results for GPT-5 Nano and Gemma 4 31B IT yet, so there is no like-for-like coding score. On overall capability, Gemma 4 31B IT leads, which tends to carry over to coding, but test on your own codebase. Both support tool calling for agent workflows.

GPT-5 Nano has the largest context window at 400,000 tokens, against 262,144 for Gemma 4 31B IT. Maximum output per response: GPT-5 Nano up to 128,000, Gemma 4 31B IT up to 32,768 tokens.

GPT-5 Nano accepts text and images; Gemma 4 31B IT accepts text and images. They handle the same number of input types.

Gemma 4 31B IT publishes its weights and can be self-hosted; GPT-5 Nano is proprietary.

Gemma 4 31B IT is the newest, released Apr 2, 2026. GPT-5 Nano came out Aug 7, 2025. Knowledge cutoff: GPT-5 Nano May 30, 2024.

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.