Gemma 4 26B A4B vs LFM3-1.7B
Google · US | Liquid AI · US · Updated June 2026
Quick verdict
Pick Gemma 4 26B A4B for fast, cheap inference from a sparse moe (3.8b active of 25.2b total) or near-31b-dense quality at a fraction of the compute and memory-bandwidth cost. Pick LFM3-1.7B for very small open-weight text model - about 1.73b parameters with a 256k-token context for on-device use or liquid says it can power agentic applications that used to need 9b parameters. On a tight budget at scale, LFM3-1.7B is the value pick.
Gemma 4 26B A4B (Google) and LFM3-1.7B (Liquid AI) are two of the models people most often weigh against each other in 2026. Gemma 4 26B A4B is an Apache-2.0 open MoE with 25.2B total but only 3.8B active parameters, delivering near-31B-dense quality at a fraction of the inference cost. LFM3-1.7B is liquid AI's October 27, 2026 on-device text model - about 1.73B parameters and 256K context, with open weights under a license capped at $10M annual revenue. They diverge most on price and context window — each quantified below from the models' real specs.
Key differences at a glance
- ▸Context window: 256K vs 256K tokens — within a few percent of each other, so treat this as a tie and test on your own long inputs, since usable recall varies by model.
- ▸Recency: LFM3-1.7B is the newer model by about 7 months (released October 27, 2026), usually meaning fresher training data and capabilities.
Side-by-side specs
| Spec | Gemma 4 26B A4B | LFM3-1.7B |
|---|---|---|
| Provider | Google (US) | Liquid AI (US) |
| Released | April 2, 2026 | October 27, 2026 |
| Context window | 256K (~393 pages) | 256K tokens (~384 pages) |
| Price (in/out) | $0.12/$0.37 per 1M tokens | Open weight (self-host / free) |
| Open weight? | Yes — self-hostable | Yes — self-hostable |
| Modalities | text, image, video, code | text |
| SWE-Bench Verified | Not published | Not published |
| MRCR v2 @ 1M | Not published | Not published |
Who wins what
Fast, cheap inference from a sparse MoE (3.8B active of 25.2B total)
Gemma 4 26B A4B
Gemma 4 26B A4B lists fast, cheap inference from a sparse MoE (3.8B active of 25.2B total) among its strengths; LFM3-1.7B does not.
Near-31B-dense quality at a fraction of the compute and memory-bandwidth cost
Gemma 4 26B A4B
Gemma 4 26B A4B lists near-31B-dense quality at a fraction of the compute and memory-bandwidth cost among its strengths; LFM3-1.7B does not.
Strong reasoning and coding (88.3% AIME 2026 no-tools, 77.1% LiveCodeBench v6)
Gemma 4 26B A4B
LFM3-1.7B is comparatively weak here — liquid says it is not recommended for agentic coding or complex multi-step reasoning because of its size
Very small open-weight text model - about 1.73B parameters with a 256K-token context for on-device use
LFM3-1.7B
Gemma 4 26B A4B is comparatively weak here — all 25.2B parameters must be loaded into memory even though only 3.8B are active per token
Liquid says it can power agentic applications that used to need 9B parameters
LFM3-1.7B
Liquid AI's October 27, 2026 on-device text model - about 1.73B parameters and 256K context, with open weights under a license capped at $10M annual revenue — and it is the newer of the two.
Part of a six-model LFM3 family covering text, vision, audio, omni and decision tasks
LFM3-1.7B
Gemma 4 26B A4B is comparatively weak here — 256K context trails 1M-token frontier rivals, and this variant has no audio input (audio is E2B/E4B/12B only)
Lowest cost at scale
LFM3-1.7B
Its weights are open, so at volume you pay for your own hardware instead of Gemma 4 26B A4B's $0.12/$0.37 per 1M tokens.
Which should you pick?
A cost-sensitive startup shipping high volume
→ LFM3-1.7B
At Open weight (self-host / free) it undercuts Gemma 4 26B A4B, and on millions of tokens that margin decides the monthly bill.
Someone analysing very long documents or codebases
→ Gemma 4 26B A4B
Larger 256K window fits more in one prompt.
Anyone whose priority is fast, cheap inference from a sparse moe (3.8b active of 25.2b total)
→ Gemma 4 26B A4B
It is specifically built for that.
Anyone whose priority is very small open-weight text model - about 1.73b parameters with a 256k-token context for on-device use
→ LFM3-1.7B
That is its strongest area.
Gemma 4 26B A4B: where it fits
An Apache-2.0 open MoE with 25.2B total but only 3.8B active parameters, delivering near-31B-dense quality at a fraction of the inference cost. Released April 2, 2026 by Google, it is built for fast, cheap inference from a sparse MoE (3.8B active of 25.2B total), near-31B-dense quality at a fraction of the compute and memory-bandwidth cost, strong reasoning and coding (88.3% AIME 2026 no-tools, 77.1% LiveCodeBench v6), and multimodal input (text/image, plus video processed as frames up to 60s) with native function calling.
Its trade-offs are real: all 25.2B parameters must be loaded into memory even though only 3.8B are active per token, and 256K context trails 1M-token frontier rivals, and this variant has no audio input (audio is E2B/E4B/12B only). At $0.12 in / $0.37 out per million tokens, it sits in the budget price band.
LFM3-1.7B: where it fits
Liquid AI's October 27, 2026 on-device text model - about 1.73B parameters and 256K context, with open weights under a license capped at $10M annual revenue. Released October 27, 2026 by Liquid AI, it is built for very small open-weight text model - about 1.73B parameters with a 256K-token context for on-device use, liquid says it can power agentic applications that used to need 9B parameters, part of a six-model LFM3 family covering text, vision, audio, omni and decision tasks, and supports 16 languages.
Its trade-offs: lFM Open License v1.0 covers commercial use only below $10M annual revenue; larger firms need a paid license, liquid says it is not recommended for agentic coding or complex multi-step reasoning because of its size, text-only; vision and audio need other LFM3 checkpoints, and no published per-token API price found. As an open-weight model, its running cost is your own hardware rather than a per-token fee.
The bottom line for this matchup
Gemma 4 26B A4B and LFM3-1.7B overlap enough that the right pick depends on your specific job. LFM3-1.7B costs less per token; Gemma 4 26B A4B holds the larger context; and each leads in its own area — Gemma 4 26B A4B for fast, cheap inference from a sparse moe (3.8b active of 25.2b total), LFM3-1.7B for very small open-weight text model - about 1.73b parameters with a 256k-token context for on-device use. Rather than crowning one, run the same hard task through both once and let the results decide.
Want both Gemma 4 26B A4B and LFM3-1.7B without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.
See pricingFrequently asked questions
Is Gemma 4 26B A4B or LFM3-1.7B better for coding?
Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Gemma 4 26B A4B leans toward fast, cheap inference from a sparse moe (3.8b active of 25.2b total) while LFM3-1.7B leans toward very small open-weight text model - about 1.73b parameters with a 256k-token context for on-device use, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Gemma 4 26B A4B or LFM3-1.7B?
LFM3-1.7B is cheaper — $0.12/$0.37 per 1M tokens vs Open weight (self-host / free).
Which has the bigger context window?
Effectively neither — 256K vs 256K tokens is a difference of a few percent. Remember advertised ≠ usable: recall typically degrades before the ceiling.
Can I use both Gemma 4 26B A4B and LFM3-1.7B together?
Yes — a multi-model platform like LumiChats gives you Gemma 4 26B A4B, LFM3-1.7B and 40+ others under one ₹69/day pass (about $1/day), so you can draft with one and cross-check with the other instead of buying two subscriptions.
Which is newer, Gemma 4 26B A4B or LFM3-1.7B?
LFM3-1.7B — released October 27, 2026, about 7 months after Gemma 4 26B A4B.
Related comparisons
Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.