Model comparison
Gemini 2.5 Flash vs Tulu 3 (Tülu 3) 70B
Gemini 2.5 Flash is the stronger model overall, scoring 39.0 to 32.6 on the Noometry Index.
Last verified . 12 shared benchmarks.
Google
Allen Institute for AI (Ai2)Summary
- They share 12 benchmarks with published results for both. Gemini 2.5 Flash scores higher in 7 categories and Tulu 3 (Tülu 3) 70B in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.5 Flash leads 38.1 to 13.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.1% for Gemini 2.5 Flash and 4.4% for Tulu 3 (Tülu 3) 70B.
- Tulu 3 (Tülu 3) 70B has downloadable open weights; the other is API-only.
Side by side
| Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | |
| Noometry Index | 39.0 | 32.6 |
| Released | 2025-04-17 | 2024-11-21 |
| Weights | Proprietary | Open |
| Context window | 1.05M | — |
| Max output | 66K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 54 | 14 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Gemini 2.5 Flash: 35.5 (#210), Tulu 3 (Tülu 3) 70B: 35.4 (#211)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Coding | 1424 | 1236 |
| SWE-bench Verified (bash only) | 28.7% | — |
| Aider Polyglot | 55.1% | — |
| WeirdML | 41.9% | — |
| ALE-Bench | 661.88 | — |
Agentic & Tool Use Not comparable
Gemini 2.5 Flash: 30.6 (#67), Tulu 3 (Tülu 3) 70B: —
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| Terminal-Bench | 17.1% | — |
| Berkeley Function Calling Leaderboard | 56.2% | — |
| TheAgentCompany | 41.1% | — |
| BALROG | 33.5% | — |
| Vending-Bench 2 | 548.84 | — |
Reasoning Tulu 3 (Tülu 3) 70B leads
Gemini 2.5 Flash: 18.2 (#271), Tulu 3 (Tülu 3) 70B: 22.9 (#172)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Hard Prompts | 1422 | 1220 |
| ARC-AGI-2 | 2.5% | — |
| SimpleBench | 41.2% | — |
| Kagi LLM Benchmark | 56.8% | — |
| ARC-AGI-1 | 33.3% | — |
| CritPt | 1.1% | — |
| EnigmaEval | 2.7% | — |
| DTBench | 76.5% | — |
| LMCA | 27.5% | — |
| Epoch Capabilities Index | 143.04 | — |
| ForecastBench | 60.6 | — |
Math Gemini 2.5 Flash leads
Gemini 2.5 Flash: 38.1 (#87), Tulu 3 (Tülu 3) 70B: 13.3 (#297)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.1% | 4.4% |
| LMArena Math | 1415 | 1242 |
| Omni-MATH | 38.5% | — |
| MATH Level 5 | — | 42.7% |
| FrontierMath (Feb 2025 set) | 4.8% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Gemini 2.5 Flash leads
Gemini 2.5 Flash: 36.1 (#162), Tulu 3 (Tülu 3) 70B: 25.1 (#257)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| GPQA Diamond | — | 46.3% |
| Humanity's Last Exam | 12.1% | — |
| MMLU-Pro | 63.9% | — |
| Confabulations | 16.8% | — |
| Vectara Hallucination Rate | 7.8% | — |
| GPQA (HELM) | 39% | — |
| LMArena Expert | 1426 | — |
Multimodal Not comparable
Gemini 2.5 Flash: 41.2 (#31), Tulu 3 (Tülu 3) 70B: —
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Vision | 1254 | — |
| GeoBench | 76% | — |
| VPCT | 46.2% | — |
| SpatialViz-Bench | 36.9% | — |
Multilingual Gemini 2.5 Flash leads
Gemini 2.5 Flash: 52.5 (#83), Tulu 3 (Tülu 3) 70B: 40.1 (#217)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Non-English | 1410 | 1236 |
| LMArena Chinese | 1450 | 1249 |
| LMArena Russian | 1415 | 1246 |
| LMArena French | 1433 | — |
| LMArena German | 1419 | — |
| LMArena Japanese | 1403 | — |
| LMArena Korean | 1385 | — |
| LMArena Spanish | 1421 | — |
Instruction Following Gemini 2.5 Flash leads
Gemini 2.5 Flash: 75.9 (#50), Tulu 3 (Tülu 3) 70B: 65.0 (#222)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Instruction Following | 1405 | 1233 |
| IFEval | 89.8% | — |
Long Context Gemini 2.5 Flash leads
Gemini 2.5 Flash: 47.5 (#16), Tulu 3 (Tülu 3) 70B: 37.1 (#217)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Longer Query | 1419 | 1224 |
| Fiction.LiveBench | 77.8% | — |
Writing & Preference Gemini 2.5 Flash leads
Gemini 2.5 Flash: 54.0 (#153), Tulu 3 (Tülu 3) 70B: 45.8 (#218)
| Benchmark | Gemini 2.5 Flash | Tulu 3 (Tülu 3) 70B |
|---|---|---|
| LMArena Text | 1417 | 1256 |
| LMArena Creative Writing | 1400 | 1231 |
| LMArena Multi-Turn | 1408 | 1252 |
| Short-Story Creative Writing | 76.5% | — |
| EQ-Bench Creative Writing | 1137 | — |
| WildBench | 81.7% | — |
Frequently asked questions
Is Gemini 2.5 Flash better than Tulu 3 (Tülu 3) 70B?
Gemini 2.5 Flash is the stronger model overall, scoring 39.0 to 32.6 on the Noometry Index.
Is Gemini 2.5 Flash or Tulu 3 (Tülu 3) 70B better for coding?
They score almost the same on coding (35.5 vs 35.4); test both on your own repository before choosing.
How many benchmarks do Gemini 2.5 Flash and Tulu 3 (Tülu 3) 70B share?
12 benchmarks have published results for both models. Gemini 2.5 Flash has 54 scored results on Noometry and Tulu 3 (Tülu 3) 70B has 14.