GPT-6 Luna vs Gemma 4 31B IT

GPT-6 Luna is the better all-round choice, scoring 77/100 against Gemma 4 31B IT (61). It leads on capability, inputs & features and context window. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond, OTIS Mock AIME 2024–2025 and SimpleQA Verified), because GPT-6 Luna has no Capabilities Index score yet.

GPT-6 Luna is cheaper at $0.10 input / $0.50 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 $0.20 per million tokens for GPT-6 Luna versus $0.205 for Gemma 4 31B IT (1× as much).

Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond, OTIS Mock AIME 2024–2025 and SimpleQA Verified): GPT-6 Luna 76.9% and Gemma 4 31B IT 53.2%. On individual benchmarks: GPQA Diamond — GPT-6 Luna 90.5%, Gemma 4 31B IT 75.8%; OTIS Mock AIME 2024–2025 — GPT-6 Luna 98.9%, Gemma 4 31B IT 73.3%; SimpleQA Verified — GPT-6 Luna 41.4%, Gemma 4 31B IT 10.4%.

There are no published SWE-bench Verified results for GPT-6 Luna and Gemma 4 31B IT yet, so there is no like-for-like coding score. On overall capability, GPT-6 Luna leads, which tends to carry over to coding, but test on your own codebase. Both support tool calling for agent workflows.

GPT-6 Luna has the largest context window at 1,050,000 tokens, against 262,144 for Gemma 4 31B IT. Maximum output per response: GPT-6 Luna up to 128,000, Gemma 4 31B IT up to 32,768 tokens.

GPT-6 Luna accepts text, images and PDFs; Gemma 4 31B IT accepts text and images. GPT-6 Luna handles the widest range of inputs.

Gemma 4 31B IT publishes its weights and can be self-hosted; GPT-6 Luna is proprietary.

GPT-6 Luna is the newest, released Sep 22, 2026. Gemma 4 31B IT came out Apr 2, 2026. Knowledge cutoff: GPT-6 Luna May 18, 2026.

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.