Compare / Llama 3.2 3B Instruct vs GLM 5.1
Llama 3.2 3B Instruct vs GLM 5.1
Which is cheaper, which is more capable, and which does better at each kind of work, from 15 results on 3 sources that measured both models.
Llama 3.2 3B Instruct or GLM 5.1? The short answer
- Llama 3.2 3B Instruct is cheaper: $0.12 against $2.13 per million tokens, 18 times less.
- GLM 5.1 scores higher on the Artificial Analysis Intelligence Index: 26.1 against 5.7.
- On our business benchmarks, GLM 5.1 ranks higher for agents and tool use, reasoning and knowledge and writing.
At a glance
| Llama 3.2 3B Instruct | GLM 5.1 | |
|---|---|---|
| Price per million tokens | $0.050 in · $0.33 out | $1.38 in · $4.40 out |
| Blended (3 in : 1 out) | $0.12 | $2.13 |
| Cheapest host, blended | $0.12 (Parasail, BF16) | $1.48 (StreamLake, FP8) |
| Intelligence Index | 5.7 | 26.1 |
| Spring Prompt overall | – | – |
| Output speed, tokens a second | – | – |
| Context window | 131,072 tokens | 204,800 tokens |
| Weights | Open | Open |
| Developer based in | the US | China |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Llama 3.2 3B Instruct | GLM 5.1 |
|---|---|---|
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 158 of 161 | 55 of 161 |
| Professional workReal tasks from banking, consulting and law, business documents and freelance projects | not measured | 37 of 51 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 183 of 195 | 62 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 187 of 187 | 35 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | not measured | 78 of 153 |
Every shared result
| Benchmark | Metric | Llama 3.2 3B Instruct | GLM 5.1 |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,146 | 1,465 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,145 | 1,448 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,179 | 1,495 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,146 | 1,460 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,167 | 1,465 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,143 | 1,453 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 5.70 | 26.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 25.5% | 86.8% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 5.3% | 30.1% |
| Artificial Analysisreported | IFBench% of instructions | 26.2% | 76.3% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 4.3% | 73.7% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 21.1% | 97.7% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 31.0% | 9.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.36 | 0.38 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 10.4 | 61.7 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Llama 3.2 3B Instruct better than GLM 5.1?
It depends on the task. Among models available today, on our benchmarks and licensed sources, GLM 5.1 ranks higher for agents and tool use, reasoning and knowledge and writing.
Which is cheaper, Llama 3.2 3B Instruct or GLM 5.1?
Llama 3.2 3B Instruct costs $0.12 per million tokens (three input to one output), against $2.13 for GLM 5.1.
Run Llama 3.2 3B Instruct and GLM 5.1 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.