Granite 4.2 8B vs LFM2 24B A2B vs Nemotron 3.5 Lightning 30B A3B
No independent capability benchmark covers every model here yet. The score uses price, inputs & features and context window.
Price in rating: Included · Exclude price
IBM
Granite 4.2 8B
65/100- ECI—
- Price$0.06 / $0.15
- Context131K
Liquid AI
LFM2 24B A2B
60/100- ECI—
- Price$0.03 / $0.12
- Context33K
NVIDIA
Nemotron 3.5 Lightning 30B A3B
69/100- ECI—
- Price$0.06 / $0.20
- Context262K
Capability evidence is incomplete
There is not enough shared capability evidence to name an overall winner. The scores (Nemotron 3.5 Lightning 30B A3B 69/100, Granite 4.2 8B 65/100, LFM2 24B A2B 60/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceLFM2 24B A2BLFM2 24B A2B $0.052 · Granite 4.2 8B $0.083 · Nemotron 3.5 Lightning 30B A3B $0.095 per 1M tokens (3:1 blend)
- Longest contextNemotron 3.5 Lightning 30B A3BNemotron 3.5 Lightning 30B A3B 262,144 · Granite 4.2 8B 131,072 · LFM2 24B A2B 32,768 tokens
- Widest inputsSame inputsGranite 4.2 8B: Text · LFM2 24B A2B: Text · Nemotron 3.5 Lightning 30B A3B: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Granite 4.2 8B | LFM2 24B A2B | Nemotron 3.5 Lightning 30B A3B |
|---|---|---|---|---|
| Price | 50% | 97 | 100 | 94 |
| Inputs & features | 30% | 35 | 25 | 45 |
| Context window | 20% | 32 | 11 | 43 |
| Specs & price only | 100% | 65/100 | 60/100 | 69/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | |||
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.06 | $0.03 (best) | $0.06 |
| Output | $0.15 | $0.12 (best) | $0.20 |
| Cached input | — | — | — |
| Blended (3:1) | $0.083 | $0.052 (best) | $0.095 |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 7 providers | Median of 1 providers | Median of 9 providers |
| Limits | |||
| Context window | 131,072 tokens | 32,768 tokens | 262,144 tokens (best) |
| Max output | 117,964 tokens | 32,768 tokens | 262,144 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | No | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | Yes | No | Yes |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | Yes |
| Availability | |||
| Weights | Open | OpenLFM Open License v1.0 | Open |
| API model ID | — | — | nvidia/nemotron-3.5-lightning-30b-a3b |
| API providers | 7 | 1 | 15 (best) |
| Released | Aug 25, 2026 | Feb 24, 2026 | Aug 11, 2026 |
| Knowledge cutoff | — | — | Sep 2025 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
Granite 4.2 8B$0.90
LFM2 24B A2B$0.54
Nemotron 3.5 Lightning 30B A3B$1.00
Which should you choose?
Which is better: Granite 4.2 8B, LFM2 24B A2B or Nemotron 3.5 Lightning 30B A3B?
There is not enough shared capability evidence to name an overall winner. The scores (Nemotron 3.5 Lightning 30B A3B 69/100, Granite 4.2 8B 65/100, LFM2 24B A2B 60/100) compare the available specifications and price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Granite 4.2 8B, LFM2 24B A2B or Nemotron 3.5 Lightning 30B A3B?
LFM2 24B A2B is cheaper at $0.03 input / $0.12 output per million tokens (median across 1 API provider). Granite 4.2 8B costs $0.06 input / $0.15 output per million tokens (median across 7 API providers); Nemotron 3.5 Lightning 30B A3B costs $0.06 input / $0.20 output per million tokens (median across 9 API providers; free on Nvidia). At a typical mix of three input tokens to one output token, that is $0.052 per million tokens for LFM2 24B A2B versus $0.083 for Granite 4.2 8B (1.6× as much) and $0.095 for Nemotron 3.5 Lightning 30B A3B (1.8× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Granite 4.2 8B has not been scored yet, LFM2 24B A2B has not been scored yet and Nemotron 3.5 Lightning 30B A3B has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Granite 4.2 8B, LFM2 24B A2B and Nemotron 3.5 Lightning 30B A3B yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Nemotron 3.5 Lightning 30B A3B has the largest context window at 262,144 tokens, against 131,072 for Granite 4.2 8B and 32,768 for LFM2 24B A2B. Maximum output per response: Granite 4.2 8B up to 117,964, LFM2 24B A2B up to 32,768, Nemotron 3.5 Lightning 30B A3B up to 262,144 tokens.
Which can read images, PDFs, audio or video?
Granite 4.2 8B accepts text; LFM2 24B A2B accepts text; Nemotron 3.5 Lightning 30B A3B accepts text. They handle the same number of input types.
Are any of these open source?
Yes, all three publish their weights (LFM Open License v1.0), so you can self-host them.
Which is newer?
Granite 4.2 8B is the newest, released Aug 25, 2026. Nemotron 3.5 Lightning 30B A3B came out Aug 11, 2026; LFM2 24B A2B came out Feb 24, 2026. Knowledge cutoff: Nemotron 3.5 Lightning 30B A3B Sep 2025.
How do you decide the winner?
This rating uses price (50%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (30%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (20%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 18, 2026.