Gemini 3.1 Flash Lite Preview vs Llama 3.1 70B Instruct
Gemini 3.1 Flash Lite Preview leads Llama 3.1 70B Instruct on the Artificial Analysis Intelligence Index, 15.6 to 6.6, and Llama 3.1 70B Instruct costs 2.7x less per output token ($0.56 vs $1.50 per million). Updated October 22, 2026; 28 benchmarks have a result for both.
Gemini 3.1 Flash Lite Preview or Llama 3.1 70B Instruct: which to pick
Only Gemini 3.1 Flash Lite Preview has a coding score, 22.7 on the Vibe Coding Index; Llama 3.1 70B Instruct has none, so the two cannot be ranked on coding.
Pick Llama 3.1 70B Instruct for output tokens 2.7x cheaper ($0.56 vs $1.50 per million).
- Shared results
- 28
- Gemini 3.1 Flash Lite Preview only
- 51
- Llama 3.1 70B Instruct only
- 19
- Like-for-like categories
- 6 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Only Gemini 3.1 Flash Lite Preview has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Gemini 3.1 Flash Lite Preview
Gemini 3.1 Flash Lite Preview leads, 15.6 against 6.6.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Gemini 3.1 Flash Lite Preview has a AA Agentic Index result, 1.6.
Directional only
Reasoning
GPQA Diamond
No call
Only Gemini 3.1 Flash Lite Preview has a GPQA Diamond result, 81.1%.
Directional only
Math
AIME
No call
Only Gemini 3.1 Flash Lite Preview has a AIME result, 83.3%.
Directional only
Knowledge
MMLU Pro
No call
Only Gemini 3.1 Flash Lite Preview has a MMLU Pro result, 86.2%.
Directional only
Web design
Design Arena: websites
No call
Only Gemini 3.1 Flash Lite Preview has a Design Arena: websites result, 1082.
Directional only
Long documents
Listed context window
Gemini 3.1 Flash Lite Preview
Gemini 3.1 Flash Lite Preview takes 1.05M tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Llama 3.1 70B Instruct
Gemini 3.1 Flash Lite Preview $0.0010, Llama 3.1 70B Instruct $0.0008 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Gemini 3.1 Flash Lite Preview has the number, $0.18.
Directional only
At a glance
- Coding index
- Gemini 3.1 Flash Lite Preview 22.7 from 3 benchmarks, Llama 3.1 70B Instruct not benchmarked.
- Widest category
- General, 30.5 points apart on average over 16 shared benchmarks.
- Widest single gap
- LMArena Text: Japanese: Gemini 3.1 Flash Lite Preview 1413 against Llama 3.1 70B Instruct 1132, 46.8 points apart on a 0 to 100 scale.
- Output price
- Gemini 3.1 Flash Lite Preview $1.50 against Llama 3.1 70B Instruct $0.56 per million output tokens, 2.7x apart.
- Context window
- Gemini 3.1 Flash Lite Preview takes 1.05M tokens, Llama 3.1 70B Instruct 131K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Gemini 3.1 Flash Lite Preview | Bars | Llama 3.1 70B Instruct | Ties |
|---|---|---|---|---|
| Coding | 2 | 0 | ||
| Intelligence | 1 | 0 | ||
| Reasoning | 3 | 0 | ||
| Math | 2 | 0 | ||
| Knowledge | 4 | 0 | ||
| General | 16 | 0 | ||
| All shared boards | 28 | 0 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index20
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| LMArena WebDev | 1257 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| AA Coding Index | 34.7 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LiveCodeBench (Vals AI) | 80.1% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| SWE-bench Verified (Vals AI) | 62.8% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Vibe Code Bench (Vals AI) | 0.0% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Code migration (Vals AI) | 4.6% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Terminal-Bench 2.1 (Vals AI) | 34.1% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 23.2% | Only Llama 3.1 70B Instruct measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 3.0% | Only Llama 3.1 70B Instruct measured |
| WeirdML | Not measured | 9.0% | Only Llama 3.1 70B Instruct measured |
| ALE-Bench | 797.7 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Text: coding | 1400 | 1260 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: software and IT | 1413 | 1260 | Gemini 3.1 Flash Lite Preview |
| LMArena WebDev: React | 1245 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena WebDev: HTML | 1276 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena WebDev: from an image | 1336 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| SciCode | 4.6% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Neither model has a result on the other 21 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 15.6 | 6.6 | Gemini 3.1 Flash Lite Preview |
| Epoch Capabilities Index | Not measured | 125.9 | Only Llama 3.1 70B Instruct measured |
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| AA Agentic Index | 1.6 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Finance Agent (Vals AI) | 30.0% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Tax Agent Bench (Vals AI) | 3.1% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Legal Research Bench (Vals AI) | 3.4% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Excel Modeling Benchmark (Vals AI) | 8.6% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| τ²-Bench Telecom | Not measured | 15.2% | Only Llama 3.1 70B Instruct measured |
| BALROG | Not measured | 27.9% | Only Llama 3.1 70B Instruct measured |
| TheAgentCompany | Not measured | 6.9% | Only Llama 3.1 70B Instruct measured |
| Neither model has a result on the other 33 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond | 81.1% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| MMMU Pro (Vals AI) | 82.5% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 40.9% | Only Llama 3.1 70B Instruct measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 4.5% | Only Llama 3.1 70B Instruct measured |
| GPQA Diamond (Epoch AI run) | Not measured | 44.2% | Only Llama 3.1 70B Instruct measured |
| LMArena Text: hard prompts | 1407 | 1241 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: expert | 1398 | 1209 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: hard prompts in English | 1403 | 1261 | Gemini 3.1 Flash Lite Preview |
| LMArena Vision: diagrams | 1242 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: homework | 1271 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Neither model has a result on the other 12 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| AIME | 83.3% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| AA Math Index | Not measured | 4.0 | Only Llama 3.1 70B Instruct measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 4.0% | Only Llama 3.1 70B Instruct measured |
| OTIS Mock AIME | Not measured | 3.6% | Only Llama 3.1 70B Instruct measured |
| MATH level 5 | Not measured | 36.7% | Only Llama 3.1 70B Instruct measured |
| LMArena Text: math | 1428 | 1252 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: mathematical work | 1417 | 1253 | Gemini 3.1 Flash Lite Preview |
| Neither model has a result on the other 6 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index40
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LegalBench (Vals AI) | 83.8% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| CaseLaw (Vals AI) | 55.0% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| CorpFin (Vals AI) | 59.4% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| TaxEval (Vals AI) | 71.8% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| MortgageTax (Vals AI) | 68.0% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| MedCode (Vals AI) | 47.6% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| MMLU Pro (Artificial Analysis run) | Not measured | 67.6% | Only Llama 3.1 70B Instruct measured |
| MMLU | Not measured | 80.1% | Only Llama 3.1 70B Instruct measured |
| LMArena Text: science | 1425 | 1276 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: medicine and healthcare | 1428 | 1257 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: legal and government | 1418 | 1283 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: business and finance | 1389 | 1235 | Gemini 3.1 Flash Lite Preview |
| FACTS Benchmark Suite | 51.4% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Neither model has a result on the other 4 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| Design Arena: websites | 1082 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: UI components | 1069 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: game dev | 1029 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: data viz | 1045 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: 3D | 1047 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: all code categories | 1072 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: SVG | 1052 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Design Arena: ASCII art | 1167 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index160
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 49.5% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| MedScribe (Vals AI) | 63.9% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Vals Multimodal Index | 41.4% | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| IFBench | Not measured | 34.4% | Only Llama 3.1 70B Instruct measured |
| DTBench | Not measured | 60.0% | Only Llama 3.1 70B Instruct measured |
| LMCA | Not measured | 14.8% | Only Llama 3.1 70B Instruct measured |
| LMArena Text | 1416 | 1261 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: instruction following | 1377 | 1231 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: creative writing | 1401 | 1232 | Gemini 3.1 Flash Lite Preview |
| LMArena Vision | 1240 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Text: English | 1414 | 1294 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: non-English | 1410 | 1219 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Chinese | 1460 | 1215 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: German | 1430 | 1222 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Japanese | 1413 | 1132 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Korean | 1392 | 1140 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: French | 1424 | 1262 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Spanish | 1421 | 1253 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Russian | 1420 | 1234 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: Polish | 1443 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Text: multi-turn | 1417 | 1256 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: longer queries | 1394 | 1241 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: writing and language | 1408 | 1236 | Gemini 3.1 Flash Lite Preview |
| LMArena Text: entertainment and media | 1386 | 1230 | Gemini 3.1 Flash Lite Preview |
| LMArena Vision: English | 1232 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: Chinese | 1278 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: reading text | 1246 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: humor | 1266 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: captioning | 1259 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: recognizing things | 1308 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| LMArena Vision: creative writing | 1248 | Not measured | Only Gemini 3.1 Flash Lite Preview measured |
| Neither model has a result on the other 15 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Gemini 3.1 Flash Lite Preview | Llama 3.1 70B Instruct | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Llama 3.1 70B Instruct costs less
Repository review
50K input + 3K output tokens
Gemini 3.1 Flash Lite Preview costs less
Agent loop
200K input + 10K output tokens
Gemini 3.1 Flash Lite Preview costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Google Gemini 3.1 Flash Lite Preview | Meta Llama 3.1 70B Instruct |
|---|---|---|
| Maker | Meta Llama | |
| Weights | No public weights listed | Open weights |
| Released | March 3, 2026 | July 23, 2024 |
| Context window | 1.05M | 131K |
| Input $/Mtok | $0.25 | $0.56 |
| Output $/Mtok | $1.50 | $0.56 |
| Coding index | 22.7 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | $0.18 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Gemini 3.1 Flash Lite Preview better than Llama 3.1 70B Instruct for coding?
Only Gemini 3.1 Flash Lite Preview has a coding score, 22.7 on the Vibe Coding Index; Llama 3.1 70B Instruct has none, so the two cannot be ranked on coding. Pick Llama 3.1 70B Instruct for output tokens 2.7x cheaper ($0.56 vs $1.50 per million).
Which is cheaper, Gemini 3.1 Flash Lite Preview or Llama 3.1 70B Instruct?
Gemini 3.1 Flash Lite Preview costs $0.25 per million input tokens and $1.50 per million output tokens; Llama 3.1 70B Instruct costs $0.56 and $0.56. Cost per solved task is known for Gemini 3.1 Flash Lite Preview only.
Which has the bigger context window, Gemini 3.1 Flash Lite Preview or Llama 3.1 70B Instruct?
Gemini 3.1 Flash Lite Preview, with 1.05M tokens against 131K.
How many benchmarks is the Gemini 3.1 Flash Lite Preview vs Llama 3.1 70B Instruct comparison based on?
28 benchmarks have a result for both models as of October 22, 2026. Gemini 3.1 Flash Lite Preview's coding index rests on 3 independent benchmarks and Llama 3.1 70B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.