Gemini 3.8 Flash vs GPT-5.6 Sol
AutomationBench-AA AutomationBench-AA: Agents orchestrate SaaS workflows through APIs; objectives receive credit only when the task has no guardrail violation
Gemini 3.8 Flash60.9
GPT-5.6 Sol51.3
GDPval-AA (work tasks) 2.1 GDPval-AA (work tasks) 2.1: Real work tasks from many professions (reports, spreadsheets, analyses), graded against expert work
Gemini 3.8 Flash46.1
GPT-5.6 Sol46.1
τ³-Banking · AA τ³-Banking · AA: Agents retrieve banking knowledge and execute tools to resolve customer-support workflows
Gemini 3.8 Flash45.8
GPT-5.6 Sol36.5
AA Analyst Agent AA Analyst Agent: Data-analyst tasks: the model explores data with tools and reports findings
Gemini 3.8 Flash · high46.3
GPT-5.6 Sol · max47.5
ITBench SRE ITBench SRE: IT incident tasks: diagnose and fix broken systems like a site-reliability engineer
Gemini 3.8 Flash · high52.5
GPT-5.6 Sol · max56.2
Terminal-Bench 4.0 Terminal-Bench 4.0 · source setting 2: Agents use a terminal to complete executable coding, system administration and research tasks in the 4.0 suite
Gemini 3.8 Flash19.7
GPT-5.6 Sol14.6
Terminal-Bench Science 0.1 Terminal-Bench Science 0.1: Agents complete expert-authored research workflows across five scientific domains in a terminal
Gemini 3.8 Flash · high10
GPT-5.6 Sol · max22.4
AA-Briefcase 1.1 · Elo AA-Briefcase 1.1 · Elo: Agents produce professional deliverables; blind pairwise comparisons produce the source Elo rating
Gemini 3.8 Flash1,181.5
GPT-5.6 Sol1,241.9
AA-LCR (long context) 1.1 AA-LCR (long context) 1.1: Questions that need reading and connecting facts across very long documents (around 100K tokens)
Gemini 3.8 Flash84
GPT-5.6 Sol80.3
DeepSWE 1.1 DeepSWE 1.1: Agents change repository code to resolve software engineering tasks under the stated DeepSWE edition
Gemini 3.8 Flash71
GPT-5.6 Sol61.1
SWE-bench Pro (v2 full) SWE-bench Pro (v2 full): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Gemini 3.8 Flash · high94.9
GPT-5.6 Sol · xhigh95.5
SWE-bench Pro (v2 hard) SWE-bench Pro (v2 hard): Agents resolve repository issues under the network-locked v2 protocol; full and 51-task Hard results are separate
Gemini 3.8 Flash · high58.8
GPT-5.6 Sol · xhigh82.4
Furniture Assembly Furniture Assembly: Models inspect furniture assembly photos, identify incorrect builds and locate the mistaken step
Gemini 3.8 Flash · high31.7
GPT-5.6 Sol · max56.7
τ³-bench · banking · 1.0.1 τ-bench official · 3-banking-1.0.1: Customer-service agents converse with simulated users, retrieve domain knowledge and change state through tools
Gemini 3.8 Flash—
GPT-5.6 Sol · xhigh46.9
Terminal-Bench Hard (version unstated) Terminal-Bench Hard (version unstated): Agents solve difficult software, system-administration and data-processing tasks in a terminal
Gemini 3.8 Flash—
GPT-5.6 Sol62.9
EnterpriseOps-Gym-AA EnterpriseOps-Gym-AA: Agents complete stateful enterprise workflows in eight domains; final database state determines success
Gemini 3.8 Flash—
GPT-5.6 Sol · max42.9
FrontierCode 1.1 FrontierCode 1.1: Agents implement changes in real repositories, graded against maintainer-defined requirements
Gemini 3.8 Flash · high36.4
GPT-5.6 Sol—
Harvey LAB-AA 1.1 Harvey LAB-AA 1.1: Legal agents work from case documents; every rubric criterion must pass and material hallucinations receive no credit
Gemini 3.8 Flash · high0
GPT-5.6 Sol—
EBR Bench EBR Bench: Agents learn to play Earthborne Rangers across repeated playthroughs
Gemini 3.8 Flash—
GPT-5.6 Sol · max44.8
| Benchmark | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| AutomationBench-AA | 60.9 | 51.3 |
| GDPval-AA (work tasks) 2.1 | 46.1 | 46.1 |
| τ³-Banking · AA | 45.8 | 36.5 |
| AA Analyst Agent | 46.3 · high | 47.5 · max |
| ITBench SRE | 52.5 · high | 56.2 · max |
| Terminal-Bench 4.0 | 19.7 | 14.6 |
| Terminal-Bench Science 0.1 | 10 · high | 22.4 · max |
| AA-Briefcase 1.1 · Elo | 1,181.5 | 1,241.9 |
| AA-LCR (long context) 1.1 | 84 | 80.3 |
| DeepSWE 1.1 | 71 | 61.1 |
| SWE-bench Pro (v2 full) | 94.9 · high | 95.5 · xhigh |
| SWE-bench Pro (v2 hard) | 58.8 · high | 82.4 · xhigh |
| Furniture Assembly | 31.7 · high | 56.7 · max |
| τ³-bench · banking · 1.0.1 | — | 46.9 · xhigh |
| Terminal-Bench Hard (version unstated) | — | 62.9 |
| EnterpriseOps-Gym-AA | — | 42.9 · max |
| FrontierCode 1.1 | 36.4 · high | — |
| Harvey LAB-AA 1.1 | 0 · high | — |
| EBR Bench | — | 44.8 · max |
Specs
| Spec | Gemini 3.8 Flash | GPT-5.6 Sol |
|---|---|---|
| Input, per 1M | !$0.75 | $4 |
| Output, per 1M | !$3.75 | $20 |
| Cost per task? | $0.93 | $0.50 |
| Speed, tok/s | — | 85 |
| Context | 1M | 1.05M |
| Weights | Closed | Closed |
| Released | Sep 2026 | Jul 2026 |
| See images | ✓ | ✓ |
| Read PDFs | ✓ | Partial |
| Hear audio | ✓ | — |
| Watch video | ✓ | — |
| Call tools | ✓ | ✓ |
| Force a tool call | Partial | Partial |
| Return strict JSON | ✓ | ✓ |
| Search the web | ✓ | ? |
| Run code | ✓ | ? |
| Use a computer | ? | ? |
What breaks when you switch
Gemini 3.8 Flash → GPT-5.6 Sol
Nothing known.
GPT-5.6 Sol → Gemini 3.8 Flash
Context: 1.05M → 1M
Sampling parameters are ignored›
temperature, top_p and top_k are ignored now and will cause HTTP 400 in future generations; remove them from requests.
No minimal thinking level›
The minimal thinking level returns an error; avoid minimal.