Ling-2.6-flash
inclusionAI · released Apr 28, 2026 · inclusionAI/Ling-2.6-flash-int4
- Type
- Open weightsMIT License
- Params
- 107B
- Context
- 262K
about 197K words of context
Our take
Written Aug 3, 2026Ling 2.6 Flash is a 107.3-billion-parameter text model from inclusionAI with a permissive MIT licence and a 262,144-token request limit. Its coding score on the Arena leaderboard is its standout result, and it is positioned as a low-cost generalist among large downloadable models.
Pick this for coding workloads where its Arena Coding score is strongest, or for long-context text tasks up to 262,144 tokens with open-weight flexibility. Use it when you need a permissive licence and the cheapest tracked tier in its parameter class. Skip it if creative writing quality matters most, if you need multimodal input, or if you want a wide choice of providers.
The case for it
- Extremely low API pricing for its parameter scale: the cheapest tracked tier is a fraction of its own higher-priced tier on the same host.
- Coding is its standout capability on Arena leaderboards, scoring well above its own overall text rating.
- Permissive MIT licence allows commercial use, modification and redistribution.
- Measurable throughput of 78 tokens per second on the budget tier, though this is only confirmed on one host.
The case against it
- Creative writing lags its other Arena categories by a wide margin.
- Active parameter count is undisclosed; only the 107.3 billion total is stated.
- Only three tracked offers, and throughput is unmeasured on two of them including the cheapest host.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)104th of 143 · 1346.1
CodingWriting and fixing code on its own
Arena Coding97th of 143 · 1412.1
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Ling-2.6-flash for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Ling-2.6-flash placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M2 Max (38-core GPU) · 96 GB
Borderline fit on an estimated size. It leaves 0.1 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 3 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.010 in / $0.030 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $0.010 / $0.030 | 262K | not measured | Unknown | Unknown | Unknown |
| Novita AI | $0.010 / $0.030 | 262K | 99 tok/s | No | No | Confirmed |
| Novita AI | $0.10 / $0.30 | 262K | not measured | Unknown | Unknown | Unknown |
Novita AIAcross the 3 listings we hold: 1 say they do not train on prompts, 0 say they do and 2 do not say. 1 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Novita AI | ✓ | ✓ | ✓ |
| Novita AI |
Tool calling: 2 of 3 listings say yes, 1 publishes no parameter list. JSON output: 2 of 3 listings say yes, 1 publishes no parameter list. Strict schema: 2 of 3 listings say yes, 1 publishes no parameter list.
When we formed this view
Dates behind this page
Prices last checked 14h ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 1 of 3 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 2 of 3 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsMIT License, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
MIT License
Fully permissive: do anything with attribution. No patent grant, unlike Apache-2.0.
Identifiers
- Hugging Face
- inclusionAI/Ling-2.6-flash-int4
- Architecture
- Mixture of experts
- Modality record
- text->text
- Catalogue slug
- inclusionai-ling-2-6-flash