Llama 3.2 1B Instruct vs Qwen2.5 7B Instruct
Llama 3.2 1B Instruct and Qwen2.5 7B Instruct cost about the same per output token ($0.201 vs $0.20 per million). Updated October 21, 2026; 3 benchmarks have a result for both.
Llama 3.2 1B Instruct or Qwen2.5 7B Instruct: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick Llama 3.2 1B Instruct for a larger context window (60K against 33K tokens) and work measured by BALROG or GPQA Diamond (Epoch AI run), where it leads.
- Shared results
- 3
- Llama 3.2 1B Instruct only
- 35
- Qwen2.5 7B Instruct only
- 4
- Like-for-like categories
- 3 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
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
No call
Only Llama 3.2 1B Instruct has a AA Intelligence Index result, 4.8.
Directional only
Long documents
Listed context window
Llama 3.2 1B Instruct
Llama 3.2 1B Instruct takes 60K tokens against 33K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Llama 3.2 1B Instruct
Llama 3.2 1B Instruct $0.0001, Qwen2.5 7B Instruct $0.0002 per turn.
List prices
At a glance
- Coding index
- Llama 3.2 1B Instruct not benchmarked, Qwen2.5 7B Instruct not benchmarked.
- Widest category
- Reasoning, 11.6 points apart on average over 1 shared benchmark.
- Widest single gap
- GPQA Diamond (Epoch AI run): Llama 3.2 1B Instruct 23.9% against Qwen2.5 7B Instruct 35.5%, 11.6 points apart on a 0 to 100 scale.
- Output price
- Llama 3.2 1B Instruct $0.201 against Qwen2.5 7B Instruct $0.20 per million output tokens, 1.0x apart.
- Context window
- Llama 3.2 1B Instruct takes 60K tokens, Qwen2.5 7B Instruct 33K.
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 | Llama 3.2 1B Instruct | Bars | Qwen2.5 7B Instruct | Ties |
|---|---|---|---|---|
| Agentic work | 0 | 1 | ||
| Reasoning | 0 | 1 | ||
| Math | 0 | 1 | ||
| All shared boards | 0 | 3 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index4 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| LiveCodeBench (Artificial Analysis run) | 1.9% | Not measured | Only Llama 3.2 1B Instruct measured |
| Terminal-Bench Hard (Artificial Analysis run) | 0.0% | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: coding | 1070 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: software and IT | 1056 | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 35 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| AA Intelligence Index | 4.8 | Not measured | Only Llama 3.2 1B Instruct measured |
| Epoch Capabilities Index | Not measured | 118.5 | Only Qwen2.5 7B Instruct measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| τ²-Bench Telecom | 0.0% | Not measured | Only Llama 3.2 1B Instruct measured |
| BALROG | 6.6% | 7.8% | Qwen2.5 7B Instruct |
| Berkeley Function Calling Leaderboard (BFCL) | 10.8% | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 38 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 19.6% | Not measured | Only Llama 3.2 1B Instruct measured |
| Humanity's Last Exam (Artificial Analysis run) | 5.5% | Not measured | Only Llama 3.2 1B Instruct measured |
| GPQA Diamond (Epoch AI run) | 23.9% | 35.5% | Qwen2.5 7B Instruct |
| LMArena Text: hard prompts | 1044 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: expert | 1007 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: hard prompts in English | 1087 | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 16 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| AA Math Index | 0.0 | Not measured | Only Llama 3.2 1B Instruct measured |
| AIME 2025 (Artificial Analysis run) | 0.0% | Not measured | Only Llama 3.2 1B Instruct measured |
| OTIS Mock AIME | 0.6% | 2.5% | Qwen2.5 7B Instruct |
| LMArena Text: math | 1086 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: mathematical work | 1098 | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 8 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | 20.0% | Not measured | Only Llama 3.2 1B Instruct measured |
| MMLU | Not measured | 72.9% | Only Qwen2.5 7B Instruct measured |
| LMArena Text: science | 1055 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: medicine and healthcare | 1009 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: legal and government | 1071 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: business and finance | 1000 | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 12 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index16 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B Instruct | Leader |
|---|---|---|---|
| IFBench | 22.8% | Not measured | Only Llama 3.2 1B Instruct measured |
| Long context reasoning (AA-LCR) | 6.7% | Not measured | Only Llama 3.2 1B Instruct measured |
| DTBench | Not measured | 47.7% | Only Qwen2.5 7B Instruct measured |
| LMCA | Not measured | 6.4% | Only Qwen2.5 7B Instruct measured |
| LMArena Text | 1055 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: instruction following | 1031 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: creative writing | 1033 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: English | 1117 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: non-English | 973 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: Chinese | 959 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: German | 1014 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: Russian | 941 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: multi-turn | 1030 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: longer queries | 1050 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: writing and language | 1030 | Not measured | Only Llama 3.2 1B Instruct measured |
| LMArena Text: entertainment and media | 1025 | Not measured | Only Llama 3.2 1B Instruct measured |
| Neither model has a result on the other 30 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Llama 3.2 1B Instruct | Qwen2.5 7B 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 3.2 1B Instruct costs less
Repository review
50K input + 3K output tokens
Llama 3.2 1B Instruct costs less
Agent loop
200K input + 10K output tokens
Llama 3.2 1B Instruct 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 | Meta Llama 3.2 1B Instruct | Qwen2.5 7B Instruct |
|---|---|---|
| Maker | Meta Llama | Qwen |
| Weights | Open weights | Open weights |
| Released | September 25, 2024 | October 16, 2024 |
| Context window | 60K | 33K |
| Input $/Mtok | $0.027 | $0.10 |
| Output $/Mtok | $0.201 | $0.20 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Llama 3.2 1B Instruct better than Qwen2.5 7B Instruct for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick Llama 3.2 1B Instruct for a larger context window (60K against 33K tokens) and work measured by BALROG or GPQA Diamond (Epoch AI run), where it leads.
Which is cheaper, Llama 3.2 1B Instruct or Qwen2.5 7B Instruct?
Llama 3.2 1B Instruct costs $0.027 per million input tokens and $0.201 per million output tokens; Qwen2.5 7B Instruct costs $0.10 and $0.20. Cost per solved task is known for neither of them.
Which has the bigger context window, Llama 3.2 1B Instruct or Qwen2.5 7B Instruct?
Llama 3.2 1B Instruct, with 60K tokens against 33K.
How many benchmarks is the Llama 3.2 1B Instruct vs Qwen2.5 7B Instruct comparison based on?
3 benchmarks have a result for both models as of October 21, 2026. Llama 3.2 1B Instruct's coding index rests on 0 independent benchmarks and Qwen2.5 7B Instruct's on 0. Where only one model has a result, the page says so instead of naming a winner.