Fugu Ultra v2.0 vs Mistral Large 4
Sakana AI · Global | Mistral AI · France · Updated June 2026
Quick verdict
Pick Fugu Ultra v2.0 for sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the deepswe coding-repair benchmark or claims to outperform flagship models like gpt-6 astra and claude fable 5.1 despite its underlying orchestration pool reportedly excluding both of them. Pick Mistral Large 4 for mistral's largest model to date — 1.05t total parameters, 49b active per token (granular moe), plus a 1.6b-parameter vision encoder or trained from scratch on 3,800 nvidia grace blackwell gpus in mistral's own european data centers, across 160+ languages. On a tight budget at scale, Mistral Large 4 is the value pick.
Fugu Ultra v2.0 (Sakana AI) and Mistral Large 4 (Mistral AI) are two of the models people most often weigh against each other in 2026. Fugu Ultra v2.0 is sakana AI's flagship multi-agent orchestration model, claiming best-or-joint-best results on 5 of 8 self-reported benchmarks against monolithic frontier models — $5/$30 per million tokens. Mistral Large 4 is mistral's October 6, 2026 public preview of 'Le Chonk' — a 1.05T-parameter, natively multimodal MoE model with open weights planned by month's end. Their biggest split is price, and the breakdown below shows exactly how that plays out for your workload.
Key differences at a glance
- ▸Price: Mistral Large 4 is about 7.4× cheaper on input ($0.68/$2.09 per 1M tokens vs $5/$30 per 1M tokens) — a large enough gap that at scale it can be the single biggest line item in the decision.
- ▸Context window: both advertise 1M tokens (~1,500 pages). Tie on paper — test on your own long inputs, since usable recall varies by model.
- ▸Recency: Mistral Large 4 is the newer model by about 26 days (released October 6, 2026), usually meaning fresher training data and capabilities.
Side-by-side specs
| Spec | Fugu Ultra v2.0 | Mistral Large 4 |
|---|---|---|
| Provider | Sakana AI (Global) | Mistral AI (France) |
| Released | September 10, 2026 | October 6, 2026 |
| Context window | 1M tokens (~1,500 pages) | 1M tokens (~1,500 pages) |
| Price (in/out) | $5/$30 per 1M tokens | $0.68/$2.09 per 1M tokens |
| Open weight? | No — API only | No — API only |
| Modalities | text | text, image, code |
| SWE-Bench Verified | Not published | Not published |
| MRCR v2 @ 1M | Not published | Not published |
Who wins what
Sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the DeepSWE coding-repair benchmark
Fugu Ultra v2.0
Mistral Large 4 is comparatively weak here — scores 61.7% on DeepSWE v1.1, trailing DeepSeek V4.1 Flash (74.2%) and Kimi K3 (68.0%) on that coding benchmark
Claims to outperform flagship models like GPT-6 Astra and Claude Fable 5.1 despite its underlying orchestration pool reportedly excluding both of them
Fugu Ultra v2.0
Mistral Large 4 is comparatively weak here — weights not yet shipped as of mid-October 2026 — API-only for now despite the open-weight framing
1M-token context window
Fugu Ultra v2.0
Fugu Ultra v2.0 lists 1M-token context window among its strengths; Mistral Large 4 does not.
Mistral's largest model to date — 1.05T total parameters, 49B active per token (granular MoE), plus a 1.6B-parameter vision encoder
Mistral Large 4
Fugu Ultra v2.0 is comparatively weak here — not a single trained model — Sakana describes Fugu as "a Multi-Agent System, Delivered as One Model" that dynamically routes tasks across a pool of other models, so head-to-head benchmark framing against monolithic frontier models should be read skeptically
Trained from scratch on 3,800 Nvidia Grace Blackwell GPUs in Mistral's own European data centers, across 160+ languages
Mistral Large 4
Mistral's October 6, 2026 public preview of 'Le Chonk' — a 1.05T-parameter, natively multimodal MoE model with open weights planned by month's end — and it runs cheaper at $0.68/$2.09 per 1M tokens.
Hybrid instruct/reasoning design — answers quickly by default or switches into step-by-step reasoning
Mistral Large 4
Mistral's October 6, 2026 public preview of 'Le Chonk' — a 1.05T-parameter, natively multimodal MoE model with open weights planned by month's end — and it is the newer of the two.
Lowest cost at scale
Mistral Large 4
At $0.68/$2.09 per 1M tokens, it is the cheaper of the two — the gap dominates the bill on high-volume workloads.
Which should you pick?
A cost-sensitive startup shipping high volume
→ Mistral Large 4
At $0.68/$2.09 per 1M tokens it undercuts Fugu Ultra v2.0, and on millions of tokens that margin decides the monthly bill.
Anyone whose priority is sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the deepswe coding-repair benchmark
→ Fugu Ultra v2.0
It is specifically built for that.
Anyone whose priority is mistral's largest model to date — 1.05t total parameters, 49b active per token (granular moe), plus a 1.6b-parameter vision encoder
→ Mistral Large 4
That is its strongest area.
Fugu Ultra v2.0: where it fits
Sakana AI's flagship multi-agent orchestration model, claiming best-or-joint-best results on 5 of 8 self-reported benchmarks against monolithic frontier models — $5/$30 per million tokens. Released September 10, 2026 by Sakana AI, it is built for sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the DeepSWE coding-repair benchmark, claims to outperform flagship models like GPT-6 Astra and Claude Fable 5.1 despite its underlying orchestration pool reportedly excluding both of them, and 1M-token context window.
Its trade-offs are real: not a single trained model — Sakana describes Fugu as "a Multi-Agent System, Delivered as One Model" that dynamically routes tasks across a pool of other models, so head-to-head benchmark framing against monolithic frontier models should be read skeptically, pricing rises to roughly $10/$45 per million tokens for prompts above 272K tokens, and benchmark comparisons against GPT-6 Astra and Claude Fable 5.1 are Sakana's own self-reported figures, not independently verified head-to-head scores. At $5 in / $30 out per million tokens, it sits in the premium price band.
Mistral Large 4: where it fits
Mistral's October 6, 2026 public preview of 'Le Chonk' — a 1.05T-parameter, natively multimodal MoE model with open weights planned by month's end. Released October 6, 2026 by Mistral AI, it is built for mistral's largest model to date — 1.05T total parameters, 49B active per token (granular MoE), plus a 1.6B-parameter vision encoder, trained from scratch on 3,800 Nvidia Grace Blackwell GPUs in Mistral's own European data centers, across 160+ languages, hybrid instruct/reasoning design — answers quickly by default or switches into step-by-step reasoning, and open weights planned by end of October 2026, which would make it among the largest open-weight models available.
Its trade-offs: still a public preview, not a finished release — Mistral says training is ongoing with 'no signs of saturation', weights not yet shipped as of mid-October 2026 — API-only for now despite the open-weight framing, launch pricing ($0.68/$2.09 per million) is a preview discount — Mistral's own model page lists a $1.36/$4.18 list price once the promotion ends, and scores 61.7% on DeepSWE v1.1, trailing DeepSeek V4.1 Flash (74.2%) and Kimi K3 (68.0%) on that coding benchmark. At $0.68 in / $2.09 out per million tokens, it sits in the budget price band.
The bottom line for this matchup
Fugu Ultra v2.0 and Mistral Large 4 overlap enough that the right pick depends on your specific job. Mistral Large 4 costs less per token; and each leads in its own area — Fugu Ultra v2.0 for sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the deepswe coding-repair benchmark, Mistral Large 4 for mistral's largest model to date — 1.05t total parameters, 49b active per token (granular moe), plus a 1.6b-parameter vision encoder. Rather than crowning one, run the same hard task through both once and let the results decide.
Want both Fugu Ultra v2.0 and Mistral Large 4 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 Fugu Ultra v2.0 or Mistral Large 4 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, Fugu Ultra v2.0 leans toward sakana's own evaluation reports best-or-joint-best results on 5 of 8 benchmarks tested, including a 74.3 on the deepswe coding-repair benchmark while Mistral Large 4 leans toward mistral's largest model to date — 1.05t total parameters, 49b active per token (granular moe), plus a 1.6b-parameter vision encoder, and that positioning usually predicts which feels better on your codebase.
Which is cheaper, Fugu Ultra v2.0 or Mistral Large 4?
Mistral Large 4 is cheaper — $5/$30 per 1M tokens vs $0.68/$2.09 per 1M tokens, roughly 7.4× apart on input.
Which has the bigger context window?
Both advertise 1M tokens (~1,500 pages). Remember advertised ≠ usable: recall typically degrades before the ceiling.
Can I use both Fugu Ultra v2.0 and Mistral Large 4 together?
Yes — a multi-model platform like LumiChats gives you Fugu Ultra v2.0, Mistral Large 4 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, Fugu Ultra v2.0 or Mistral Large 4?
Mistral Large 4 — released October 6, 2026, about 26 days after Fugu Ultra v2.0.
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.